Heater INSIDE thermostat?

Can someone explain the use of a heater INSIDE a mechanical (bimetal strip) room thermostat? (It usually just seems to be a resistor placed near the bi-metallic strip).

All it seems to do for me is make the thermostat very inaccurate. If the temperature is only a bit lower outside than inside, the temperature inside is what I set the stat to. As the temperature outside drops further, the inside temperature strays further and further from what I set it to (in the colder direction). This is presumably because whenever the heating is on, the stat thinks it's warmer than it really is, because of the internal heater. If I disconnect this internal heater, the stats functions properly.

I've replaced the offending pile of crap with a digital one anyway, but I wanted to know why they design them like this.

Reply to
Lieutenant Scott
Loading thread data ...

It's an anticipator, and should be set so as to introduce exactly the right amount of heat inside the thermostat casing so that it doesn't have to wait for the heat to percolate from a radiator across the other side of the room, thereby tailoring the response so that it stabilises sooner.

It's a bit like trying to throw a Yo-Yo away whilst still holding onto the string. And yes it's a crap plumbers idea.

DerekG

Reply to
DerekG

It is to make the thermostat/heating more efficient by allowing the stat to 'predict' a rise/fall in temperature. If the stat is in good condition, it does work better than a plain on/off stat, where the temperature can fall/rise a lot more than one supplied with a neutral. It wil also save you money by turning off/on the boiler more efficiently.

Alan.

Reply to
A.Lee

It isnt. I've been on a training course that shows the graphs of heat and gas used. The heated stats use less gas, and keep the room at a steadier temperature than the standard on/off stats.

Alan.

Reply to
A.Lee

I've found that the convection of the room air from the heat source (a radiator) heats the room very evenly anyway. The stat switches off when the room reaches the required temperature without that bloody heating thingy!

As you said, it just doesn't work as designed - it's guessing! All that happens is when the heating is needed to be on more, it thinks it's going to be warmer than it is, so it doesn't switch it on enough.

Reply to
Lieutenant Scott

Seems to work a LOT better without the "heater" connected. It simply turns on a degree below what I set it to, and off at the temperature I set it to.

Reply to
Lieutenant Scott

room thermostat? (It usually just seems to be a resistor placed near the bi-metallic strip).

The bimetal strip will have a fair measure of hysteresis, i.e. it might switch on your heating at 18C, but not switch it off until 22C, so the room will swing between these two.

The accelerator heater works to reduce this hysteresis. In the example above, if the resistor is chosen to heat the bimetal strip by 3C, it will reduce the hysteresis of the thermostat from 4C down to 1C. Additionally, it adds a bit of a time lag, so that even if the hysteresis is reduced to nothing (which might lead to rapid switching), the lag extends both the on and off times a bit to reduce rapid switching, which is not a good idea for many types of heating, or the life of the contacts if it's switching a large load.

temperature is only a bit lower outside than inside, the temperature inside is what I set the stat to. As the temperature outside drops further, the inside temperature strays further and further from what I set it to (in the colder direction). This is presumably because whenever the heating is on, the stat thinks it's warmer than it really is, because of the internal heater. If I disconnect this internal heater, the stats functions properly.

I have found cases where the supplyu and load are connected the wrong way around, which can really screw up thermostats with accelerator heaters.

Reply to
Andrew Gabriel

room thermostat? (It usually just seems to be a resistor placed near the bi-metallic strip).

Mine seems to be about 2C. It's getting replaced with a digital one anyway.

I've found that bi-metal strips simply can't switch some loads, like my central heating. There must be a lot of inductance in the load or something - it's just a motorised valve. As it switches (I think it's more at the "off" switching) an arc forms on the gap in the bi-metal strip, which obviously heats it and causes it to stay on. This makes a worrying noise and interferes with radio and TV reception.

temperature is only a bit lower outside than inside, the temperature inside is what I set the stat to. As the temperature outside drops further, the inside temperature strays further and further from what I set it to (in the colder direction). This is presumably because whenever the heating is on, the stat thinks it's warmer than it really is, because of the internal heater. If I disconnect this internal heater, the stats functions properly.

Presumably you mean swapped supply live and switched live - which would make the internal heater stay on continuously. I would think this would make it like a stat without a heater, but with the temperature shunted off by a few degrees.

Reply to
Lieutenant Scott

room thermostat? (It usually just seems to be a resistor placed near the bi-metallic strip).

temperature is only a bit lower outside than inside, the temperature inside is what I set the stat to. As the temperature outside drops further, the inside temperature strays further and further from what I set it to (in the colder direction). This is presumably because whenever the heating is on, the stat thinks it's warmer than it really is, because of the internal heater. If I disconnect this internal heater, the stats functions properly.

wanted to know why they design them like this.

It reduces the hysteresis of the system by forcing it to shut off the heat earlier than it would do naturally.

Hysteresis is a bit like mechanical backlash when a gear chain goes into reverse.

Reply to
Graham.

A plumber could not come up up with such a clever idea :')

It was a widely used idea to prevent over reaction of the heating system, due to the wide hysteresis of a mechanical thermostat in the days before modern electronic stats. The stat switches on, which then powers a resistor, which warms the bi-metal sensing element, so the stat gets both the warmth of the heating system and the warmth from the resistor.

Reply to
Harry Bloomfield

room thermostat? (It usually just seems to be a resistor placed near the bi-metallic strip).

temperature is only a bit lower outside than inside, the temperature inside is what I set the stat to. As the temperature outside drops further, the inside temperature strays further and further from what I set it to (in the colder direction). This is presumably because whenever the heating is on, the stat thinks it's warmer than it really is, because of the internal heater. If I disconnect this internal heater, the stats functions properly.

wanted to know why they design them like this.

I know about hysteresis in electronics, but you've lost me with the gears :-)

Reply to
Lieutenant Scott

But it's wildly inaccurate. And there seems to be no adjustment for them.

Reply to
Lieutenant Scott

Or another way to look at it is it narrows the thermostat's hysteresis, ie makes the off-on temperature closer to the on-off point.

Reply to
Tim Watts

It's a very cheap way to improve the responsiveness of the system. I once disconnected one, and found out that re-connecting it it made a noticeable improvement to my comfort in the room with no noticeable increase in heating costs.

It was a single small room with a blown air calor gas powered heating system, but the same principle applies to a central heating system as used in my house.

Reply to
John Williamson

The trouble is it makes the on-off point drift depending on how much the heating is used.

Reply to
Lieutenant Scott

I think the resistor in this one is making too much heat. It never switches the heating on for very long, and I don't think it's good for a central heating system to be cycled on and off repeatedly. I much prefer it to come on for an hour at a time.

Reply to
Lieutenant Scott

Imagine the chain has a lot of slack.

Reply to
Tim Streater

Ahhhh, Sturmey Archer gears.

Reply to
Lieutenant Scott

The thermostat is meant to measure the air temperature. The heater causes= room air to pass through the thermostat housing by natural convection. The= re is otherwise no reason why the room air should be in contact with the th= ermostat element, the element would react to a change in room temperature a= t some indeterminable time. It introduces an application error, but the err= or is predictable. The only other way of getting an air flow through the th= ermostat housing would involve a fan.=20

The other reason is that the heater switches off the heating before the he= ating reaches the thermostat set-point but, because of the residual heat in= the radiators and pipes, the room temperature will continue to rise slight= ly after the switch off.

It can be effective apparently, but most temperature sensors have been he= ading in the thermistor direction for a while. You very rarely see the heat= er connected because it requires the use of another neutral wire.

Reply to
Onetap

room thermostat? (It usually just seems to be a resistor placed near the bi-metallic strip).

temperature is only a bit lower outside than inside, the temperature inside is what I set the stat to. As the temperature outside drops further, the inside temperature strays further and further from what I set it to (in the colder direction). This is presumably because whenever the heating is on, the stat thinks it's warmer than it really is, because of the internal heater. If I disconnect this internal heater, the stats functions properly.

wanted to know why they design them like this.

I first came across the concept when studying electronic servicing specifically magnetic recording, the B-H curve, hysteresis loops and all that crap.

It wasn't until I noticed a similar effect on mechanical systems like the slow motion vernier tuning arrangement on my home-made transmitters that it sank in.

You turn the dial one way and the frequency changes smoothly. Turn it the other way and nothing much happens for a small amount of rotation because the gears aren't tightly meshed, then smooth operation returns albeit with a calibration error compared with the other direction.

I found the mechanical backlash analogy very useful in understanding the magnetic effect and wished the lecturer had used it.

Reply to
Graham.

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.