Boiler short cycling with low flow temperatures

[MrPotts]

11 minutes ago, Nickfromwales said:

So all the way around the gardens and back to basics then? KISS lol. 👌

When I started the journey a few years ago all I wanted was to replace my old analogue Honeywell room stat with a new digital room stat. I then succumbed to the clever marketing and online youtubers extoling the virtues of Hive, Wiser, Tado, Nest etc only to spend the next 2 years and 3 winters trying to get the heating to work how I wanted. I've now given up and gone back to a single room thermostat albeit the Drayton Wiser room thermostat.

Edited February 5 by MrPotts

[marshian]

2 hours ago, seanblee said:

 

I increased by half a turn this morning and another quarter this afternoon (leaving some lockshields now wide open) and it hasn't made any real difference to warm-up this afternoon. The setpoint was 18 for the lounge during the day, increasing to 20 at 5pm - Evohome decides when to start warm-up so it hits this temperature at this time, and it looks like it kicked off just after 3pm today:

Yes it’s still cycling but the important bit to consider is if the boiler still overshooting the set point with more open rads? 

[seanblee]

57 minutes ago, marshian said:

Yes it’s still cycling but the important bit to consider is if the boiler still overshooting the set point with more open rads? 

Unfortunately yes. I've increased the set point to 65 and it seems better, but I'll know more tomorrow morning - right now, the rooms have overshot their set temperature, so I think Evohome needs to re-adapt.

[marshian]

1 minute ago, seanblee said:

Unfortunately yes. I've increased the set point to 65 and it seems better, but I'll know more tomorrow morning - right now, the rooms have overshot their set temperature, so I think Evohome needs to re-adapt.

 

I'm not concerned about rooms over shooting

 

Evohome can do what it likes but if you still have a situation where the flow temp spikes when the boiler shortly restart overshoots and shuts down quicky then clearly the circuit still has a restriction to flow that the boiler is not happy with.

 

You also say the system is noisy - now normally that would indicate a strong flow thro the circuit and I wouldn't be looking there but now I am.

 

Let me show you my issue when for a reason I didn't understand I had to wind up the pump to speed 2 in order to stop the boiler short cycling due to temp overshoot. This massively increased system noise

 

Now admittedly the pump had been in there 15 years and there was a magnaclean installed a couple of years before I found this but it was installed after the pump.........

 

The system noise kept SWMBO awake which is sub optimal and also annoyed me as the system had always been whisper quiet on the lowest pump speed

 

Cleaned out (and there was a lot of stuff in the vanes you can't see) I was able to drop back to speed 1 and the short cycling stopped

 

Not say this is your problem but whilst Evohome is almost certainly cycling the boiler the over shoots aren't down to Evohome

 

[marshian]

One more thing - has the boiler been range rated down for the CH side of the boiler?

 

 

Edited February 5 by marshian
Add link to how too

[marshian]

Went googling - installation manual was rubbish, user manual equally so that you tube video gave me a big clue

 

Turns out all the interesting set up stuff you can do with boiler parameters

 

range rating

Anti cycle

Pump speeds and modes

Pump over run time

 

Greenstar_CDi_Classic_Engineer_Service_booklet.pdf

 

 

[seanblee]

Thanks, yes I've played with most of that. I range rated it down to the minimum (40%) for heating (installers left it on 100%). I tried decreasing the anti cycle time to 1 minute but I've bumped it back up to 3 now. On the pump, the installers left it in the default of proportional high, which was very noisy, so I dropped it back to proportional low a few weeks after it was fitted, and now it's on fixed speed pump step 2. I only got monitoring working through HA in the past few months so I've no idea if it's always behaved like this and I've just not known, or if this is something new.

 

I can try pulling the pump and seeing what state it's in, it's internal to the boiler though so I'm not sure how easy that's going to be. I was wondering if it maybe needed a flush.

[marshian]

3 minutes ago, seanblee said:

Thanks, yes I've played with most of that. I range rated it down to the minimum (40%) for heating (installers left it on 100%).

 

It's a 30kW boiler for CH so 40% of 30kW is 12 kW - do you really need 12kW to cover your heat loss and allow a little for a reasonable recovery  time - Im surprised you can't range rate it down to minimum - the manual seems to indicate that you can range rate to the minimum??

 

3 minutes ago, seanblee said:

I tried decreasing the anti cycle time to 1 minute but I've bumped it back up to 3 now.

 

Anti cycle time - decreasing it makes no sense - longer is what you want

 

It's the time the boiler waits out before restarting

 

On some boilers once the flow temp and return temp get down to 10 deg below set point they fire anyway

 

I'd also look at the Anti-cycle flow temperature differential and consider increasing that to give a longer coast period via a lower restart temp.

 

I don't know your boiler like I know Glow worm but GW/V pump over run gets first call followed by ACT so I'd start with adjusting PORT - it didn't have the Anti-cycle flow temperature differential adjustment so never got to try that

 

3 minutes ago, seanblee said:

On the pump, the installers left it in the default of proportional high, which was very noisy, so I dropped it back to proportional low a few weeks after it was fitted, and now it's on fixed speed pump step 2. I only got monitoring working through HA in the past few months so I've no idea if it's always behaved like this and I've just not known, or if this is something new.

 

OK I did wonder if it's always been like this or something new.

 

3 minutes ago, seanblee said:

I can try pulling the pump and seeing what state it's in, it's internal to the boiler though so I'm not sure how easy that's going to be. I was wondering if it maybe needed a flush.

 

It's only 4 allen bolts to take the pump assembly off the body (oh and an isolate and drain - you don't need to take the pump out of the boiler - just remove the motor and impellor from the pump body

[seanblee]

Good call, I'm not sure why I thought 40% was minimum, I've dropped it now to 28%. I reduced the anti cycle time when I was struggling to hit the target room temperature in the lounge - I've bumped it up to 5 mins and increased the hysteresis to -15K. Behaviour on this boiler is that if the flow overshoots, both the anti cycle time and the hysteresis have to be satisfied before it re-fires. I've also increased the pump overrun from 5 to 10 minutes. Thanks again for your continued input, much appreciated!

[marshian]

37 minutes ago, seanblee said:

Good call, I'm not sure why I thought 40% was minimum, I've dropped it now to 28%.

 

Much better 11.2 kW - will it go down to the absolute min?

 

37 minutes ago, seanblee said:

I reduced the anti cycle time when I was struggling to hit the target room temperature in the lounge - I've bumped it up to 5 mins and increased the hysteresis to -15K.

 

Different ways of doing things - room struggling to reach target temp - my first reaction is to increase flow thro that rad

 

ACT probably pays to be a bit mathmatical

 

ACT of 5 mins means a max number of fires in 60 mins is 60 / 5 = 12 now of course the burn time is going to reduce the number of cycles but even if the burn time is only 1 min each time it'll drop the number of cycles to 10 per hour so idealy you want a burn time of 5 mins or more to get below 6 cycles per hour.

 

If you aren't achieving that then consider increasing the ACT

 

Hysterisis 15 deg might help - boilers only tend to get fussy when the delta between flow and return goes over 20 deg

 

37 minutes ago, seanblee said:

Behaviour on this boiler is that if the flow overshoots, both the anti cycle time and the hysteresis have to be satisfied before it re-fires.

 

OK it's handy to have that understanding

 

37 minutes ago, seanblee said:

I've also increased the pump overrun from 5 to 10 minutes.

 

Normally I'd say do one thing at a time but in your case - throw everything at it to see what happens - it's an easy step to reset to factory settings

 

37 minutes ago, seanblee said:

Thanks again for your continued input, much appreciated!

 

Not a problem at all

 

If I could comment some of this detail would have been good to have known in the first post of the thread - yes it would have made it a bit wordy but it might have saved a bit of time.

 

What I really am interested in is if the purge and initial burn is still at an elevated boiler kW output......

 

My current boiler hits the circuit with 60% of the boiler max for 90 secs - it range rated to 5.86 kWh - so it's still smashing ~10kW at the circuit for the first 90 secs

 

My old glow worm was 75% of max for 30 secs trouble is that was 18 kW - pretty damn hard for the circuit to deal with that if the overshoot is only 14 deg away

[seanblee]

This morning is looking rather better: 

 

Perhaps more interesting is the burner power - this was yesterday morning: 

 

 

And this is today: 

 

Pretty dramatic difference! I might still raise the ACT a little to reduce the cycles when the room temperature is closer to the set point, but this is the first morning I can remember where I came downstairs to find the lounge at the target temperature, not 1-2 degrees below.

 

28% is as low as it'll go for range rating, the register won't accept a smaller number. It looks like it very briefly fires at a higher power level but almost immediately drops back to the configured setting.

Edited February 6 by seanblee
Typo

[marshian]

That’s better - still probably things can be tweaked but I’ll take that as a win

[marshian]

Important thing (stating the bloody obvious) is the boiler has to get the house warm - no point having a boiler thats tweaked to the max to give longest burns possible at lowest flow temps if the house is cold……….

 

I reckon you’ll be able to reduce the flow temps along with a few more tweaks and maybe even drop the pump speed too

 

last question are the lockshields reset to where they were or still a 1/2 turn open more?

[seanblee]

25 minutes ago, marshian said:

last question are the lockshields reset to where they were or still a 1/2 turn open more?

They're all back where they were, so approx 10 degrees drop across each radiator at 65 degrees flow. At this point, I think the changes that have made a difference are eliminating (most of) Evohome's TRVs, range rating, tweaking anti cycle and increasing the flow temperature. I'll give it a few days then I might try lowering the flow a few degrees in tandem with opening the lock shields a little and see what that does, but I'm happy it's at a good starting point now. Thanks again!

[marshian]

My approach would be

 

1. lower pump speed (I hate system noise and if you are running 24/7 with setbacks it’s costing you money via higher wattage drawn by the pump)

 

if boiler overshoots 

 

2. open lockshields a bit more but with 9 or so rads in circuit you really shouldn’t need a high pump speed and the balancing should be fairly easy to achieve

 

Then with a stable system circuit running quietly I’d start to back down the flow temp - as you do this your individual rad deltas between flow and return will drop anyway so don’t be tempted to tighten the circuit to keep the same 10 deg delta 

 

heat geek had a really good table of what to expect delta wise at the boiler at various flow temps - it’s a really useful table that stops you from fretting about low delta T at the boiler.

 

As you drive the flow temp down you might need to adjust the boiler hysterisis but you are doing that to be inline (ish) with the circuit flow and return temps (to get to a point where the boiler doesn’t spend ages coasting while the circuit cools)

 

all of the above with an eye on keeping the house comfortable temp without long recovery times after a setback at night

 

 

[marshian]

On 06/02/2025 at 14:52, marshian said:

heat geek had a really good table of what to expect delta wise at the boiler at various flow temps - it’s a really useful table that stops you from fretting about low delta T at the boiler.

 

Found it

 

https://www.heatgeek.com/do-we-really-need-dt20/

 

Basically as you lower the flow temps the delta between flow and return will drop

 

I'm currently running ~30 deg Boiler flow with a 7 Deg drop at the return of the boiler

 

Flow Temp	Return Temp	Diff
80	56.0	24.0
75	52.5	22.5
70	49.0	21.0
65	45.5	19.5
60	42.0	18.0
55	38.5	16.5
50	35.0	15.0
45	31.5	13.5
40	28.0	12.0
35	24.5	10.5
30	21.0	9.0

[Marvin]

Hi @seanblee

 

I have not read all the thread but I would like to add that one of the things I did was to turn down the flow rate of the pump and balance the radiators to avoid cycling problems. 

 

If you have microbore you definitely need something like a Magna clean. After any plumbing pipe work it is good to flush the entire system out during summer. 

[John Carroll]

8 hours ago, marshian said:

 

Found it

 

https://www.heatgeek.com/do-we-really-need-dt20/

 

Basically as you lower the flow temps the delta between flow and return will drop

 

I'm currently running ~30 deg Boiler flow with a 7 Deg drop at the return of the boiler

 

Flow Temp	Return Temp	Diff
80	56.0	24.0
75	52.5	22.5
70	49.0	21.0
65	45.5	19.5
60	42.0	18.0
55	38.5	16.5
50	35.0	15.0
45	31.5	13.5
40	28.0	12.0
35	24.5	10.5
30	21.0	9.0

 

Is this a heatgeek table, if so, or whoever,  then they must be changing the flowrate to get those sort of dTs?.

If you use the third row, 70C/49C/21C, as the base, this gives a flowrate of 0.502LPM/kW

and (assuming the same flowrate) a dT of 11.22C (vs 15.0C) at 50C flowtemp and a dT of 2.88C (vs 9.0C)  at 30C flowtemp. (Assuming a 20C Room Temp)

 

 

 

Edited February 11 by John Carroll

[marshian]

28 minutes ago, John Carroll said:

 

Is this a heatgeek table, if so, or whoever,  then they must be changing the flowrate to get those sort of dTs?.

If you use the third row, 70C/49C/21C, as the base, this gives a flowrate of 0.502LPM/kW

and (assuming the same flowrate) a dT of 11.22C (vs 15.0C) at 50C flowtemp and a dT of 2.88C (vs 9.0C)  at 30C flowtemp. (Assuming a 20C Room Temp)

 

 

 

 

Table generated by "Rule of thumb"

 

Heat Geek article said

 

As the controller causes the return temperature to drop and get ever closer to the room temperature, it becomes increasingly more difficult to shed the heat. This is because the 'heat transfer coefficient' of the radiators drops. For example, if you have a room of 20°c you'll never be able to get a DT of 20 if your flow temperature is for example 30°c. A more realistic target is a DT that's around 30% of the flow temperature.

 

For example; If we have a flow temperature of 70°c, (70 x 0.3) gives a DT of 21°c. If your flow temperature is 50°c this would give a DT of 15°c (50 X 0.3) and so on. This is not exact, it's just to get the flow rate in the right ballpark. There are more complex calculations for this if you need but really not worth it.

[SimonD]

11 hours ago, John Carroll said:

Is this a heatgeek table, if so, or whoever,  then they must be changing the flowrate to get those sort of dTs?.

 

yes, they must be, but like most Heatgeek blogs, the nuance of all factors required to get a heating system working are missing, giving what I think is a misleadingly simplistic account, even if the principles are fairly simple.

 

A lot of the figures above simply won't work very well with radiator systems without impractical modifications.

[marshian]

2 minutes ago, SimonD said:

 

yes, they must be, but like most Heatgeek blogs, the nuance of all factors required to get a heating system working are missing, giving what I think is a misleadingly simplistic account, even if the principles are fairly simple.

 

A lot of the figures above simply won't work very well with radiator systems without impractical modifications.

 

Sheesh.......................

 

I posted it because it's useful to have an expectation of what your return temps are likely to be based on a lowered flow temp.

 

30 Deg Boiler Flow Temp

20 Deg Room Temp

You aren't going to get a DT of 20 and a boiler return temp of 10 Deg

 

I think you need to consider the audience that the blogs are aimed at - mainly the general public / DIY enthusiasts

 

It's a "rule of thumb" - a "guide" call it whatever you like it's not "Gravity"

 

They do training sessions for Heat Engineers that are way more detailed

[SimonD]

2 hours ago, marshian said:

 

Sheesh.......................

 

I posted it because it's useful to have an expectation of what your return temps are likely to be based on a lowered flow temp.

 

30 Deg Boiler Flow Temp

20 Deg Room Temp

You aren't going to get a DT of 20 and a boiler return temp of 10 Deg

 

I think you need to consider the audience that the blogs are aimed at - mainly the general public / DIY enthusiasts

 

It's a "rule of thumb" - a "guide" call it whatever you like it's not "Gravity"

 

They do training sessions for Heat Engineers that are way more detailed

 

Yes, but is it useful to believe its all down to delta T, even for general public and enthusiasts without also understanding the implications this has on heat output and mass flow rate and how this impacts the function of the system as a whole?

 

By definition a lower delta T distributes less heat and therefore requires higher flow rates for the same output, some of which are impractical. And then with radiators heat transfer is based on Mean Water Temp to Air Temp differential as shown by the spreadsheet above so many radiator systems just won't work at low flow temperatures.

 

Yep, maybe it's a sheesh post......😉

 

 

[marshian]

@seanblee Any updates sir?

[seanblee]

Well, I'm still running at over 60 degrees flow temperature - I think I'm ultimately going to need to replace the two vertical radiators with bigger ones to get more heat into the room at lower flow temperatures. Other than that, I think things are looking pretty good - this is the flow temperature today, and the burner stats show it running at 28% from 4am for nearly 90 minutes, so it's no longer short cycling.

 

[marshian]

20 hours ago, seanblee said:

Well, I'm still running at over 60 degrees flow temperature - I think I'm ultimately going to need to replace the two vertical radiators with bigger ones to get more heat into the room at lower flow temperatures. Other than that, I think things are looking pretty good - this is the flow temperature today, and the burner stats show it running at 28% from 4am for nearly 90 minutes, so it's no longer short cycling.

 

 

Looks better

 

You can potentially avoid bigger rads by heating longer (ie 24/7 or with very minimal set backs outside of scheduled heating periods)

 

I was heating to a schedule and running 50 deg flow temps (at zero OAT) - to get the room up to temp within the schedule

 

Moved to 24/7 heating and I'm running 32 deg flow temps (at zero OAT)

 

Setbacks with recovery didn't seem to save any energy v a 24/7 heating period

 

Only proviso is the boiler needs to be happy at the lower flow temps