If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
When only 1 radiator is demanding heat from the boiler, no demand signal is sent until the room temperature is 0.5C below the set up. This causes the room temperature to oscillate.
Please can you check that you are getting a similar affect with your system, thanks.
Why do you think it is a bug? What is the background to this heat demand. What is the time and temperature profile.
Need more info to understand your question
I work for Resideo, posts are personal and my own views.
I think there is a bug because I assume a signal would be sent to the boiler when heat is needed, not waiting until the room temperature is 0.5C below the set point.
I think there is a bug because I assume a signal would be sent to the boiler when heat is needed, not waiting until the room temperature is 0.5C below the set point.
This is called a control differential. 0.5 deg is very close control. It will depend on a range of variables. That is why I asked for context.
I work for Resideo, posts are personal and my own views.
Normally this is not a problem, other radiators in the house are demanding heat so keeps the room temperature within 0.1C. Only in the evening when only the sitting room radiator is on do I get this temperature oscillation.
This is called a control differential. 0.5 deg is very close control. It will depend on a range of variables. That is why I asked for context.
I tend to agree with you about the control differential, though this thread is starting to get a bit messy as there seem to be two competing topics going on within it. I don't think a suitable explanation was ever provided as to why the graphs eric posted on page 2 seem to show evohome regulating the room temperature so poorly in comparison to a plain old room stat. However, I think mikey's posts relate mostly to a comment Rameses made back on the first page:
"The Evo will start firing the boiler at +0.5°C above Set Point - this is around the control band of Set Point ±0.5°C"
This didn't make sense to me at the time, as it seems to suggest that evo is preemptively firing the boiler, which sounds like a big waste of energy. From what you're saying, and as I suspected was probably the case all along, HR92s open and heat is demanded when the room temperature falls 0.5C below the set point. Every control system has to have some amount of hysteresis, so I'm not sure what mikey means when saying "I assume a signal would be sent to the boiler when heat is needed". Even if it was 0.00001C (and I don't even want to think about the amount of cycling that would cause), there has to be some control differential, unless the system can somehow exist in both on and off states at the same time!
Last edited by electronicsuk; 27 October 2015, 09:51 PM.
The proportional band (this is a PI control) is 1.5 degrees so below this the heat demand will be 100%. Within the proportional band the self learning fuzzy logic algorithm will know the heat up and cool down characteristics and create a heat demand. On a TPI wireless relay this will translate to a switched demand based on the minimum run time and cycle rate.
If I understand the question, on low load there may be a scenario where running the boiler for 1 minute may be too much. So evohome may decide not to as it may result in overshoot. Remember the HR92 radiator controller is modulating so that can close down to reduce the mean temperature on the radiator.
The other thing to consider is the size of the boiler and radiators. If these are oversized evohome may not have enough control authority. Evohome is very very good but it cannot do magic.
I work for Resideo, posts are personal and my own views.
Assume or another word expect a signal to be sent to the boiler when heat is needed. No signal sent when 1 radiator in operation and temperature is above set point-0.5C. eg. set point 21C only when room temperature drops to 20.5C is a signal sent to the boiler to switch on.
Assume or another word expect a signal to be sent to the boiler when heat is needed. No signal sent when 1 radiator in operation and temperature is above set point-0.5C. eg. set point 21C only when room temperature drops to 20.5C is a signal sent to the boiler to switch on.
I work for Resideo, posts are personal and my own views.
So you're saying it's normal that Evohome will let the room temperature drop 0.5 below setpoint before firing the boiler, causing pretty big temperature oscillations, because it's "very close control", even though a simple room stat does a much better job at keep a steady temperature?
The HR92 'modulating', or actually closing down, will have little effect when you're just heating 1 room by the way. The boiler pump will pump it's heat through only 1 or 2 radiators. Unless the HR92 is 95% closed or probably even more, the water WILL go through the radiators. Unless you have a bypass set VERY loose, which would mean you're wasting a lot of gas just heating up a bypass.
The story about being oversized and control authority is strange too. When heating just 1 room (which Evohome is advertised as being able to do, as a USP!), almost ANY boiler is going to be oversized. There's just no boilers sized to heat only 1 room in the house. This means the thermostat has to be able to deal with this. And a simple room stat IS able to do this almost perfectly (see my graphs). I can assure you there was no magic involved.
Not saying anything Erik. I'm still trying to understand the OPs concern.
Again just sharing information from my experience and viewpoint to try to be helpful. My opinion.
So you're saying it's normal that Evohome will let the room temperature drop 0.5 below setpoint before firing the boiler, causing pretty big temperature oscillations, because it's "very close control", even though a simple room stat does a much better job at keep a steady temperature?
The HR92 'modulating', or actually closing down, will have little effect when you're just heating 1 room by the way. The boiler pump will pump it's heat through only 1 or 2 radiators. Unless the HR92 is 95% closed or probably even more, the water WILL go through the radiators. Unless you have a bypass set VERY loose, which would mean you're wasting a lot of gas just heating up a bypass.
The story about being oversized and control authority is strange too. When heating just 1 room (which Evohome is advertised as being able to do, as a USP!), almost ANY boiler is going to be oversized. There's just no boilers sized to heat only 1 room in the house. This means the thermostat has to be able to deal with this. And a simple room stat IS able to do this almost perfectly (see my graphs). I can assure you there was no magic involved.
I work for Resideo, posts are personal and my own views.
With the TPi thermostat, are you still only sending heat to one radiator? Or is the rest of the system accepting heat from the boiler?
the former. It's actually 2 radiators, the ones in the living room.
What make and model boiler is it? What output? What is the heat loss of the room and the size of the radiator?
Vaillant VHR24-28C from about 2000. I think it's about 22KW and able to modulate to about 50%. I have the max water temperature set to 60 instead of the default 90 to save gas. It should mean less KW's going into the radiators as well. The living room radiators are about 4KW combined. I've got no idea about the heat loss. The house is from the '30s with big windows, so there's certainly loss. But it's between 2 other houses and the glass is insulated. So it's not awful... It's really all pretty standard set-up when compared to other comparable houses and their radiators/boilers.
And: other thermostats work just fine. We've tried with a really really old mercury-filled one from honeywell and a slightly newer Honeywell Round (wired, with TPI, using the same 10 minute cycles as Evohome). The BIG difference I noticed between Evohome and Honeywell Round in on/off behavior is that I NEVER saw the BDR91 every skipping a 10 minute cycle, untill at least 0.7 above setpoint (after wich it would finally stop further heating the room, and stay totally OFF for at least 2 hours, radiators becoming totally cold, the living room cooling down a full degree very uncomfortable, etc). And trust me, I've spent a lot of time monitoring that thing. Once the BDR91 turned ON once, it would continue to turn ON for the minimum run time every 10 minutes, untill the living room finally being way above setpoint. This heating up process would often take at least 1.5 hours. So the Evohome system should have plenty of time to react to this situation, but just didn't. The Honeywell Round however would skip a 10 minute cycle pretty often when the desired temperature was (about to be) reached and no additional heat was needed. So to me it's really no wonder that the Round was doing a much better job at maintaining a constant temperature. I think the BDR91 algorithm is missing a cycle-skip-part. If for example it receives only a 10% heat demand, it will still go ON for the minimum run time every 10 minute cycle. It only stays totally off when it receives a 0% demand. However, Evotouch does not seem to send a 0% demand, unless the room is probably at least 1 above setpoint.
PS: I'm using Tado now and it keeps the room at 1 temperature just fine as well. Very steady graph from my temperature logger. Now just have to wait untill next year, when they'll release their smart radiator heads for zoning.
@Dan: I gave you the info you asked for. Is there anything you can conclude from it?
Comment