Radiator balancing problems

[Securitybyte]

Wow! Thank you everyone for the invaluable tips and advise! That makes much more sense now - I'll buy an IR sensor and have another go

I had been wondering about reducing the pump speed on the boiler. It was actually going to be my next question - as the system tends to be quite noisy when only one or two zones are on (which is the case more often that not). I noticed when I had all the rads on to balance it how nice and silent it was..

The boiler and radiators are newly installed. The boiler is a Worcester Greenstar 32CDi Compact. Possible pump maps are:

0 - pump variable speed setting
1 - constant pressure low
2 - constant pressure low/middle
3 - constant pressure middle/high
4 - constant pressure high

The factory setting is 4. If I was to have to reduce it after my next balancing attempt, I guess I'll reduce to 3? Or does anyone know if I could try using the variable speed?

Regarding the automatic bypass valve, I thought I was going to have to get one of these installed to fix the noise, but according to the boiler manual, it does a bypass valve built in

Many thanks !

[DBMandrake]

Maximum pressure is unlikely to be correct. I would give settings 2 and 3 a try in conjunction with the balancing.

If you set the pump speed too low the symptoms will be one or more of the following:

* Excessive temperature drop across radiator panels, eg you measure more than 12 degrees across many of the radiators and opening the lock shield valves don't help.
* Radiators struggling to come up to temperature in a reasonable time.
* Boiler short cycling - if the flow is too low the boiler can't transfer the heat it generates to the radiators quickly enough so will be forced to keep cycling off even though the radiators may not be up to temperature. (If the flow is sufficient, if you turn the system on from cold the boiler should run continuously without cycling off until all the radiators are nearly up to temperature)

If you see any of those symptoms increase the speed one setting and try again.

If the speed is too high the main symptom is excessive noise, but it can also reduce efficiency because you don't get the desired temperature drop between flow and return that helps get the boiler into condensing mode. (Irrelevant if you have a non-condensing boiler like us)

Our pump has three speeds and I've found by trial and error that the system is happiest on the middle speed. On the lowest speed the flow isn't quite high enough to prevent short cycling and on the highest speed some radiators start gurgling and making excess noise.

[Securitybyte]

Hi Mandrake

That's great info - thank you!

I've not re-balanced the radiators yet since my new IR thermometer did not arrive until today. I've also not yet tried changing the pump speed - it is still at max.

I have been watching the boiler over the last few evenings however and observed some quite odd cycling problems.

It goes like this:

- BDR91 calls for heat
- Boiler starts
- Flow temperature quickly rises from about 49C through to 60C in about 10 seconds (this is after it's been on a while - I've not watched it from a cold start)
- Then over a period of say 15 seconds*:
- 64C flow, 56C return
- 65 flow, 56.5 return
- 66 flow, 57 return
- 67 flow, 58 return
- Status 204 - 'System waiting, primary flow temperature above setpoint'
- Burner stops
- Pump continues to run at about 60% for a few minutes, with a return of approx 50C
- After 5 or so minutes - Lather, rinse, repeat

*Sometimes it runs for longer - 5 minutes or more, but most frequently I've watched it stop after between 2 and 15 seconds.

The boiler setpoint is set to 63C - the economical temperature recommended for the boiler. I have tried turning it up but it made no difference - the flow temp still exceeded the setpoint.

During this time, the rooms are still below their set points and the BDR91 is still lit up.

Initially I thought it was because I only had one radiator on, however I've tested it with 5 radiators on this evening and it is the same story.

The LSV is fully open on two of the radiators as they are upstairs and furthest away from the boiler - these have HR92s so should have been open 100%
The third LVS is closed quite a bit otherwise it is very noisy (towel rail) - HR92 so should have been open 100%
The other two are partially closed as they are closest to the boiler - manual TRV - turned down to 1 - still enough for rads to get pretty hot


Bizarrely...

As I have been writing this, the rooms have all come up to their setpoints and the boiler has successfully completed five consecutive 'calls' from the BDR91 without stopping prematurely.
The flow was between 56C and 60C and the return between 45 and 49.5C for the whole period which the light on the BDR91 was on.

So I am a bit confused why it would be cycling on and off while the zones were coming up to temperature, when to maintain them at their temperatures it is coping just fine. If it can
modulate low enough to keep it ticking over, why can't it modulate low enough during the initial heat?

Edit - the sixth time it came on it stopped after about 15 seconds with status 204 again

Apologies for the long post!

Many thanks in advance

[killa47]

Hi Securitybyte
I sympathise with your woes as I too am struggling to stop my WB 40cdi conventional form short-cycling. WB term is on my boiler as "Gradient Limitation" - flow temperature rising too quickly.

I wonder if your problem is to do with the parameter setting of "Anti-cycle flow temperature differential" which WB normally set as 10C. So if your return temp is inside 10C of flow temp, it will turn off the burner for 2 minutes (maybe 3 minutes depending on parameter setting).

I have been confused for some time with whether to increase/decrease pump speed, open/shut LSV's, increase/reduce flow temperature, improve ABV function etc.

I have asked my original plumber to attend, service system and check if - having installed the Evohome system in November (which I think works fine) - I need to re-assess the way in which the Honeywell 2-way zone valves work. They are currently left permanently open to feed upstairs/downstairs CH.

I suspect maybe the upstairs valve which only normally feeds towel rails and a landing radiator during daylight hours returns water inside the 10C differential range.

If I turn on majority of rads in both areas (up & dwon), the boiler seems to behave. Another plumber advised my boiler is almost industrial strength so in "tickover" mode it might short cycle. Possible suggestion there was a more elaborate ABV system.

Not sure if any of the above helps you but perhaps DBMandrake might throw more light on your problem.

Ian K

[DBMandrake]

I have been watching the boiler over the last few evenings however and observed some quite odd cycling problems.

It goes like this:

- BDR91 calls for heat
- Boiler starts
- Flow temperature quickly rises from about 49C through to 60C in about 10 seconds (this is after it's been on a while - I've not watched it from a cold start)
- Then over a period of say 15 seconds*:
- 64C flow, 56C return
- 65 flow, 56.5 return
- 66 flow, 57 return
- 67 flow, 58 return
- Status 204 - 'System waiting, primary flow temperature above setpoint'
- Burner stops
- Pump continues to run at about 60% for a few minutes, with a return of approx 50C
- After 5 or so minutes - Lather, rinse, repeat

*Sometimes it runs for longer - 5 minutes or more, but most frequently I've watched it stop after between 2 and 15 seconds.

The boiler setpoint is set to 63C - the economical temperature recommended for the boiler. I have tried turning it up but it made no difference - the flow temp still exceeded the setpoint.

During this time, the rooms are still below their set points and the BDR91 is still lit up.

Initially I thought it was because I only had one radiator on, however I've tested it with 5 radiators on this evening and it is the same story.

If the temperature is rising very quickly it means either (a) the pump speed is far too slow (not the case in your situation as the pump is on maximum) or (b) the radiators are hardly open (or maybe not at all) and most or all the flow is returning via the automatic bypass loop - which will cause a very fast temperature rise indeed, as you have a ~ 22KW boiler heating no more than about 10 litres of water circulating through the heat exchanger, the bypass valve and back again through a short run of pipe.

When you say you had 5 radiators "on" what was the relationship between the set point and the measure temperatures of each zone ? Because it's a proportional system you'll find that once a room is up to temperature the radiator valve opening can be quite small to maintain a steady temperature in the room, so the flow through the radiator may be only a trickle. You can check the valve opening by holding the button on the HR92 until the configuration menu comes up then turning to option 10 and setting that to 1 - it will then show the valve position as a percentage from 0-100%.

Different valve bodies have different opening points - so for example just because it says the pin is 20% out doesn't necessarily mean any water is flowing as the valve may not start flowing until the pin is 30% of the way out. So on one particular valve it might just start to open at 30% and be fully open by 70% for example. Another valve may not start flowing until it's 40% open and yet be fully flowing at 60%.

The HR92 doesn't know precisely at what point the valve opens or closes, all it does is increases the opening if it is consistently below temperature and decreases the opening if its consistently above temperature. Over time it will find the right opening to get equilibrium in the room temperature, however because it doesn't know exactly where the valves are opening in some conditions it might be calling for heat for a while when the valve is not actually open enough to allow any water to flow. (And it has no real way to know that is happening except that the measured temperature is not rising as it is expecting it to, so gradually it will keep opening the valve bit by bit until the temperature does start to rise)

Bizarrely...

As I have been writing this, the rooms have all come up to their setpoints and the boiler has successfully completed five consecutive 'calls' from the BDR91 without stopping prematurely.
The flow was between 56C and 60C and the return between 45 and 49.5C for the whole period which the light on the BDR91 was on.

Sounds like at this point the radiator valves were sufficiently open to provide enough load on the boiler. In the scenario I painted above because the rooms remained persistently below their setpoints the HR92's gradually increases their valve opening until they were actually flowing sufficiently. They may have been calling for heat prematurely perhaps due to valves that need to be open a lot before they start flowing.

So I am a bit confused why it would be cycling on and off while the zones were coming up to temperature, when to maintain them at their temperatures it is coping just fine. If it can
modulate low enough to keep it ticking over, why can't it modulate low enough during the initial heat?

When you say coming up to temperature, how many degrees below the set points were the zones ? The honeywell system has a proportional band of about +/- 1.5 degrees, so if any zone is at least 1.5 degrees too cold the boiler will fire at 100% duty cycle and the HR92 itself will be more or less fully open. However if all zones are within their +/- 1.5 degree proportional bands relative to their setpoints, the boiler will generally not be fired constantly by the BDR91, and the HR92's will be partially open.

Does your boiler support Opentherm ? If it does you're probably better off using the Opentherm bridge instead of the BDR91 as it will give the Honeywell controller direct control over the flow temperature - instead of cycling the BDR91 off and on under duty cycle control to try to reduce the boiler output when demand is low (which works but is somewhat crude and inaccurate) it will directly command the flow temperature to drop on the boiler to the desired flow temperature.

It may even help with the symptoms you're seeing, however I'm not sure that I would necessarily worry about what you're seeing - at the point where demand is quite light and radiator valves are mostly closed you probably will see a bit of short cycling of the boiler due to high flow through the automatic bypass valve, because of your boiler trying to be perhaps a bit too "clever", however as soon as demand increases a bit the boiler will stop short cycling.

[DBMandrake]

Hi Securitybyte
I sympathise with your woes as I too am struggling to stop my WB 40cdi conventional form short-cycling. WB term is on my boiler as "Gradient Limitation" - flow temperature rising too quickly.

I wonder if your problem is to do with the parameter setting of "Anti-cycle flow temperature differential" which WB normally set as 10C. So if your return temp is inside 10C of flow temp, it will turn off the burner for 2 minutes (maybe 3 minutes depending on parameter setting).

I have been confused for some time with whether to increase/decrease pump speed, open/shut LSV's, increase/reduce flow temperature, improve ABV function etc.

Sounds like the same scenario as Securitybyte - assuming your pump speed is not set too low flow temperature rising "too quickly" is going to be because the radiator valves are largely closed and most of the flow is returning via the automatic bypass valve, or possibly your automatic bypass valve is set too low. What differential pressure is it set to ?

To some degree I think this is inevitable on the Honeywell system if you have all radiators controlled (eg no bypass radiator left open) because under some conditions it will run the boiler when the HR92's are hardly open, but when it does that it generally runs the boiler at a low duty cycle. (Eg it may be turning it on for 2 minutes out of 10 minutes) But that may still be enough to trigger the "Gradient Limitation" warning on your boiler during that 2 minute period.

I have an ancient conventional boiler that has no electronics at all apart from a retrofitted digital flow temperature controller I added to replace the original mechanical flow stat - and I see the same rapid rise of flow temperature when for example there are only 2 zones in the house active and they are already up to temperature - but my boiler doesn't care, the flow temperature controller just shuts it off when it reaches the the target temperature and doesn't complain about the rapid rise.

The issue here really is that your boiler is being too clever for its own good - a fast flow temperature rise when all the radiators are nearly closed and the ABV is flowing heavily is to be expected. Does it actually cause a lockout situation where you have to manually reset it to get things working again, or does it just complain on the screen but continue working normally ? If the latter I probably wouldn't worry about it to be honest, if it does prevent it working I would see if there was any way to relax the "Gradient Limitation" threshold so that it doesn't get tripped.

[killa47]

Hi DBMandrake

As always, an excellent explanation of some confusing (to a layman) operational features of our boiler systems.

You have at the least reassured me that the gradient limitation is not something I should worry too much over. I t has done this pretty much since the Evohome was put in but at no point as it triggered a lockout.

Equally, the boiler does not complain noise-wise nor do I have audible symptoms throughout the system. In fact, since Evohome, I enjoy having the visual control on the CH/DHW in the house.

I did wonder about the ABV setting but this is just about the only part of the system I am unfamiliar with. I know they can be adjusted and I have a Center EHE0200281 ABV. It is adjustable and I have the install guide so I will look further at this. I suspect I might need to ask my plumber to make sure it is correctly rated for the flow rate/pump head.

Your suggestion to relax the gradient limit also occurred to me but I cannot see a parameter for this when I checked them yesterday. Maybe WB can advise me if this is user configurable.

Your comments are much appreciated, including those re Securitybyte, and I will digest these further.

Many thanks

Ian K.

[Securitybyte]

Hi Ian

Thanks for the information. Makes me a bit happier to know it is not just me! My Anti-fast cycle flow temperature is set to the default of 6 degrees K. I assumed that was Kelvin but that would equate to -267.15 degrees C, so not sure how that works... Anti-fast cycle time is on the default of 5 mins.

Your issue sounds pretty similar - all ok with lots of radiators on..

I've been doing some more monitoring and have narrowed it down a bit more, and will explain this in my next post to Mandrake!

Thanks again

[DBMandrake]

Hi Ian
My Anti-fast cycle flow temperature is set to the default of 6 degrees K. I assumed that was Kelvin but that would equate to -267.15 degrees C, so not sure how that works...

6 degrees will be a differential temperature, so in that regard Kelvin and Celsius are equivalent. (Unlike Fahrenheit where the units themselves are different by a 9/5 ratio, not just a different zero starting point)

[Securitybyte]

Hi Mandrake

Thank you again for your great reply.

When the 5 radiators were on, the two manual ones were on number 1 but got hot pretty quickly, so I think the pins were quite well open. The Evohome TRVs were all atleast a few degrees below their setpoints, so should have been fully open but I didn't check. The light was lit up on the BCR91 the whole time. I thought the valves would have been open more fully during the time it was heating up, and thus a lower return temperature. When it was up to temperature, with the valves more closed, I'd expect the return temperature to be potentially higher and thus then get the status 204. But it was the other way around.

Opentherm sounds like it would do exactly as is needed, but unfortunately my boiler does not appear to support it

It doesn't cause a lock out, it just gives the status and goes quieter as the burner stops. After 5 minutes or so it resumes. The Anti-fast cycle flow temperature is configurable between 2 and 15 degrees K. It's currently on 6 (Default) but I don't understand enough about it to change it at the moment..


I balanced the radiators today. I need to do some further tweaks but most of the radiators have a drop of 9-12 degrees C. I managed to achieve this with the pump speed on low - all the radiators got fully hot, and I am pretty sure the boiler did not cycle at all over the couple of hours I was doing it. So I was quite happy with that. Afterwards though, after cooling it down and switching it all back on again - the radiators were barely hot. Strange as they were fine as I was balancing them and I double checked them afterwards. I've since re-opened them all back up by half a turn and will check what this has done to the temp drops a bit later.

Back to running the heating with one radiator on... The zone started at 21C (hotter than normal due to the balancing). I requested 25C and the TRV claimed to be 100% open. Very shortly after the boiler starting, the flow had climbed to 73C with a return of 33C. The boiler shut off with the 204 status.

I made the following changes to see if they would have any affect, but none of them made any difference at all:

- increased the pump speed to 3
- put the pump speed back to maximum
- reduced the maximum heating output to its minimum of 7kW - just to test
- turned up some radiators with manual TRVs

No luck - it keeps switching off early due to the high flow temperature when the zone is still below the setpoint. I've been monitoring it since 2000. At least for the last hour it has barely stayed on for more than 10 seconds, with the zone 1/2 a degree below the setpoint the whole time.

I understand why it is doing it, as you kindly explained, but I won't be able to rest until it stops doing it

The last time it did it, it said the return was only 47.5C, yet the flow went right up to 70C before it switched off. So if my observations are correct, sometimes the flow temperature exceeds the setpoint because the return temperature is high. And sometimes it is because there is not enough water flow through the heat exchanger? I've turned the pump back onto maximum, but I guess if some TRVs have since closed now.. But isn't that was the automatic bypass is for? Sorry, a lot of these questions are just me thinking out loud and I'm not expecting answers to all of them.

I've noticed another issue too. If the heating is running and I switch on another radiator in an unused room, it sends through a load of cold water to the boiler which triggers blocking error 349 - "CH: Boil Detect, large delta T despite minimum burner load". This is where the boiler is "operating at minimum burner load and there is a greater than 18C Flow & Return temperature difference". What a stupid design.

Many thanks