So as a further update, I also asked Jurian if the wavy lines correspond to the valve position, or to heater demand. He’s saying this:
Personally, I don’t know a thing about PID, except for the Wikipedia article is just Googled. But... maybe one of you guys know more?
This seems to me to be consistent.
To put it simply, a pid controller, if properly tuned, does much better than an on/off controller. It drastically reduces overheating and underheating. I notice that as the days go by, the system adapts itself better and better to the heating demands.
There are percentages next to the wavy lines on the Care and Protect page. I assume that you are talking about the relationship between the wavy lines and these percentages. As far as I can make out these percentages represent demand rather than controller output. The demand signal is modified by PID functions to give the controller output which is what controls the valve.
Not much more than two weeks ago I put in a suggestion that Self Tuning PI Controllers (or indeed PID) should be used - combined with improved temperature measurement. This suggestion has already been moved to "The Big Idea Archive" That was a bit quick! Perhaps because they already have a form of PID control which they are happy with or perhaps I should have not been too specific regarding the suggested solution.
I assume and also it would be logical that the Tado PID controllers are on their servers - following the Tado goal of being Hardware Agnostic.
My archived suggestion was intended in particular to suggest that the controllers should be "Self Tuning". I would be interested to know if they are. A PID controller is not necessarily self tuning. If a controller is correctly tuned, any fluctuating +ve and -ve deviations of the measured value from the set point will noticeably diminish over a short period. You should be able to see this effect an a graph.
The bottom line is that it is not easy for the controllers to adapt to many different systems. For instance half of my system is microbore so some of my controls would be in saturation for lengthy periods (i.e. controller output at 0 or 100%).
Finally, proper control is impossible without reliable and accurate temperature measurement and vertically mounted SRT's (with built in temperature sensors) fitted close to the radiator definitely do not give that.
There are percentages next to the wavy lines on the Care and Protect page. I assume that you are talking about the relationship between the wavy lines and these percentages. As far as I can make out these percentages represent demand rather than controller output. If the percentages are controller outputs there must be some other limit on how often the valve actuators can move as they do not move every time the percentages change.
Not much more than two weeks ago I put in a suggestion that Self Tuning PI Controllers (or STPID) should be used - combined with improved temperature measurement. This suggestion has already been moved to "The Big Idea Archive" That was a bit quick! Perhaps because they already have a form of PID control that they are happy with or perhaps I should have not been too specific regarding the suggested solution.
I assume and also it would be logical that the Tado PID controllers are on their servers - following the Tado goal of being Hardware Agnostic.
My archived suggestion was intended in particular to suggest that the controllers should be "Self Tuning". I would be interested to know if they are. A PID controller is not necessarily self tuning. If a controller is correctly tuned, any fluctuating +ve and -ve deviations of the measured value from the set point will noticeably diminish over a short period. You should be able to see this effect on a graph.
From what I can see, there is some room for improvement in both temp measurement and control.
The bottom line is that it is not easy for the controllers to adapt to many different systems. For instance half of my system is microbore so some of my controls (compared with in other rooms) would be in saturation for lengthy periods (i.e. controller output at 0 or 100%).
However many steps the TADO TRVs are capable of opening and closing in I can tell you they are far more effective than some of their rivals. As a recent TADO convert its a pleasure to hear the valves open, and feel the radiators beginning to warm, moments after a call for heat rather than the ~ 20 minutes previously tolerated . . .
So I asked some additional questions to Jurian (who's really quick to respond, by the way) concerning the relationship between tado modulating the radiator valves and modulating the heating request from the heater. He told me both operate in around 5% steps, and are related but not necessarily the same.
He did mention that when only 1 tado valve requests power, it will always open 100% and modulate the required heat only. When 2 or more valves request power, both will modulate.
Maybe that explains the somehow contradictory explanations so far. So valves will open/close only when 1 valve requests heat. If more do so, both valves and heating request will modulate apparently.
We learn a little step by step...
My reference to "demand" was actually meant to refer to the error signal or PID controller input rather than to the PID controller output signal. Despite all the various posts and speculations on the subject I am still no wiser as to precisely what the percentages on "Care & Protect" actually represent.
Due to the fact that Tado users are on multiple different configurations (relay, Opentherm, combi, heat only etc) we are not always comparing like with like if we get too specific. My system is a relay heat only Y plan system with vented hot water tank and Tado extension kit. If Opentherm is an option for me I might consider it, but am probably more likely to go for the weather compensation system that is available for my boiler (a Viessmann) as I can then have some influence on which weather compensation control ramp is in use.
For me even with only one SRT in use it seems that it can be partially open and I need to have another radiator online with no SRT to ensure that the boiler is not locked out by low flow. Perhaps that is what applies to my particular set up...
I read this thread with interest as one of my radiators is constantly overshooting by 2 or 3 degrees. It is the radiator in the kitchen, which is almost next to the heater. Before I had Tado that radiator would be open about 25% all the time to prevent it from heating too much and returning too hot water back to the heater. I suspect it's now opened too much before it overshoots.
Is there a way to limit the opening of a Tado valve to combat this?
@Schippie you mention you limit the max flow. That sounds like something that could help me. How do you do that?
[Quote]"He did mention that when only 1 tado valve requests power, it will always open 100% and modulate the required heat only. When 2 or more valves request power, both will modulate."[/Quote]
In my recent experience this is, unfortunately, absolutely untrue. I've had a single radiator calling for heat at 23% (whatever that means exactly), boiler on, and zero flow through the radiator. The valve simply was not open enough to allow water to pass. I do have a bypass rad, so that gets 95% of the heat while the room that actually wants heating gets almost none, or actually none.
Indeed I am extremely suspicious that battery preservation had taken precedence over optional functioning of the heating. Far too often I see only one or two radiators calling for heat, but with only modest valve openings. Well my boiler cannot output less than 9kW, so with only one or two radiators sipping water instead of glugging it down the boiler shuts off prematurely as target flow temperature (set by ebus) is reached before the radiators have even seen any of it. I've seen burn times as short as 30 seconds on occasion, followed by the inevitable five minute pump overrun as a thimble full of tepid water is pointlessly circulated round the system. How is that a sensible way to run a system?
If the radiators are balanced it is even worse, with radiators choked off at both ends. There is no point that I can see in having partial valve openings in these (or any other) circumstances. If the radiator needs heat then have at it. The boiler will thank you.
@johnbur I do have my suspicions that, with valves as responsive as those controlled by Tado, the benefits of balancing are questionable. For the most part it is generally the case (in my house) that only a few radiators request heat at any one time. This rest are actually shut off and not bleeding off heat, leaving plenty for those that need it.
Also, my heating is on 24x7, since the house is usually occupied. Therefore the heating doesn't have to play catch up, as it might in the evening for a house unoccupied during the day. So when heat is needed in a room it will not take long before it is up to temperature, so once again heat is available to other radiators soon enough.
With this in mind, although I did go to the trouble of balancing my system, I have actually backed off some of the lockshield valves following observation of operation of Tado. That said, I do believe that Tado has improved on the partial opening performance of the valves, reducing this silly trickle nonsense, so I think it is possible to be more relaxed about the lockshield. But I would say better a little more open than a little more closed.
My one Tado TRV has been causing the heating on to be far longer than needed. That radiator asks for a small amount of heat but Tado doesn't let any into the radiator so the radiator stays cold so the only way the room is warm is by accident from the rest of the house 🙄 Rest of the house has been turning into a sauna.
I played around with the pin and made sure it was free. Took off the tado, instant heat. I remounted it and it was back to cold. I have the vertical mounting ones so I unscrewed the base mount half a turn. I think this forces a higher minimum flow as the pin can't be pushed so far down. Now there is some warmth at all. I think this is the issue that depending on the exact nature of the pin you have to fiddle with the amount it is screwed on or off to the radiator to fine tune how much you get. If you screw it further down to absolute rock bottom it's turning the radiator off off (or does in this case) If you unscrew it a bit it can't turn it off off so you don't get the absurd condition of the radiator that is wanting heat is effectively turned off.
@srichards Try talking directly to Tado Support. I found myself in the same situation with one of my rads, calling for heat, boiler on, no heat through the radiator due to closed Tado valve. I think they did take my feedback on board as the problem did disappear. Whether that was a general tweak to firmware globally for everyone or a little tweak just for me I do not know, but worth a shout. There is no guarantee they will see your comments here and do anything.
I already have talked to them. They say it's operating normally! Heating is stuck on now. Room is just below the set temperature and that radiator is basically off. It's in the slightly warmer weather that leads to edge case heating that is no good. Previously it remained off and the house was too cold instead as it would drop up to 0.5c below set temperature before coming on. I can't decide which is worse.
I've tweaked the base another small amount. Let more heat in again to try and compensate as the radiator needs a small amount of heat whenever the heating is actually on as it's usually the one wanting heat.
Regarding this "He did mention that when only 1 tado valve requests power, it will always open 100% and modulate the required heat only. When 2 or more valves request power, both will modulate".
There seems to have been a change to the way this works. When only one radiator in the house is on and that radiator is at a low level heat request, I can hear a farily loud 'hissing' sound as the water is forced though the partially open valve. This noise goes away at high heat request levels, when the valve is more open, and also goeas away (or at least gets quieter to the point I can barely hear it) when multiple radiator valves are open. I also can hear the bypass valve opening slightly so some of the water is not even going through the radiator.
I dont remember this being the case in the past, the heating has only recently been turned on after the summer so it has been a while but I'm sure I would have noticed the sound before.
This doesn't bother me too much, the radiator still gets hot and the lower flow though the radiator might prevent overshoots, but I just thought it was an interesting change.
I've been thinking about this for a while ... noticed we have cold rooms and when the radiator calls for heat the rad seems nowhere near as warm as I'd expect it to be. Others in the house that are calling for heat seem much warmer (and yes, they have been bled).
I do wonder if this means rooms can take longer to heat than necessary due to restricted flow?
It doesn't. In a properly balanced system many/most radiators will already be restricted by the lockshield valve. When the radiator requests heat it should be supplied with heat from a fully open flow. There is no earthly point in adding further restriction. If only one radiator is requesting heat and Tado blocks the flow then all you do is heat the bypass instead of the room that wants the heat. It is mental.
I've spent a year trying to understand the Tado system and learning how to combat the stupid default operation. I'm pretty much there, but it should just work out of the box without trial and error by the customer.
My solution is to make most of my radiator valves independent and to set their temperature a degree or two above the temperature target of the radiator(s) which do talk to the zone controller. In this way the independent rads are almost always open for business, with unrestricted flow, so that they can contribute to warming the WHOLE HOUSE with an expanded emitter surface area. This is much more effective than trying to heat the house with a single bypass radiator or, worse, integrated bypass valve within the boiler. But even though the valves may be wide open (much of the time), there is an upper ceiling imposed to stop things getting out of hand. And by actually discharging the heat through multiple rads it helps keep return temperatures low and improves condensing boiler efficiency
Tado, if heat is required then open the bloody valves!
I struggle to understand why anyone would think the valve would be anything other than fully open or fully closed. Why would it be? It’s like expecting a conventional TRV to heat a room up quicker if it’s set to 5 instead of to 3. It doesn’t! 5 or 3 just sets the temperature at which the valve closes.
Theres an update from 2021 half way down this page from admin (Rob) about a future change that deliberately causes the partly open radiator issue we are having here. I can see how it helps the people in that thread but by not making this change a user choice they've made things better for some heating systems at the expense of others
At this point I'm beginning to think of looking for alternatives and ditching the whole system