Hi folks
I’ve been uploading my gas boiler data from Home Assistant for a while and I can see all of the data arriving in Emoncms - app view . However, on a few days I’m missing the summary bar graph. e.g. below. What is causing the gaps? I’ve checked and can’t see any missing info on the uploaded data graphs. Any advice would be very welcome. Cheers.
Tim R
Hello @gr0mit I’ve just fixed the energy feeds for you, they needed recording using “log to feed (join)”, this joins across small gaps in the data. The missing bars result from null values in the kWh data at midnight, log to feed join removes these
Hi Trystan - Amazing! Thanks so much. How does my gas boiler performance compare with others? I am currently running it with a home-built weather compensation curve controlled by Home Assistant. Found a way to programatically set the flow temp of my Vaillant system boiler, and I’ve also reconfigured to PDHW (you can see the higher temp setting when CH is not called for, where DHW heating the tank.
Cheers
Tim
Hello @gr0mit Im sure you are already aware of @tomasmcguinness’s monitored gas boiler? My gas boiler data is live!
Looks like yours is working well!
There’s also @critictidier @showpiece.bsky.social on Bluesky and Emoncms - app view
I can see a lot of misfires which isn’t great. If your boiler supports OpenTherm then you might benefit from GitHub - Alexwijn/SAT: The Smart Autotune Thermostat (SAT) is a custom component for Home Assistant that works with many compatible gateways for your boiler. @gr0mit
Hi - Thanks for your observations. Can you explain what you mean by misfires?
Just FYI my control system is using ebusd and Home Assistant to set the flow temp of the CH water to match a heat curve that I’ve implemented based on external temperature and required internal setpoint of 21 Celsius. The heat curve is based on Michael dePodesta’s Vaillant Ecotherm equation.
I looked a while back at buying the Dutch Vaillant Opentherm board but it was nearly €200 after import duties (stupid stupid Brexit) so I opted for the ebusd interface instead, and implemented everything in HA. Because why not re-invent it?!
Cheers
Tim
Also, part of the problem is that the boiler, a Vaillant 24kW beast of a thing, doesn’t modulate down low enough, so it’s always cycling. Any way around this?
What’s your boiler model?
Hi
It’s a Vaillant Ecotec Plus 624
GC No. 41-044-47
Was installed in 2011.
Cheers
Tim
Misfires are when boiler starts but then immediately ends the cycle < 1min.
It seems that your boiler modulates down to about 8kW, not great but with low load compensation it would help for sure.
I can’t advise and promise anything but I guess you would need this module
https://www.vanwalraven.com/en/artnr/336472/vaillant-opentherm-modul-vr33
And then you’d need to buy the OpenTherm gateway from the Nodo Shop
For information about what low load is
Low load - in simple terms - max cycles per hour (3 or 4 for a boiler) where the duration of each cycle is calculated and controlled with precision (like PWM). Uncontrolled / normal cycling is when they are not restricted on how many per hour and duration is not precisely controlled.
I have a Glowworm Ultracom 18hxi (was 30hxi). This is basically the same boiler as your Ecotec, both based on a Saunier-Duval design, and had much the same problems as you. I also have it under eBUS control. It has some “interesting” features
There are a few things you can do
My boiler graphs are here
You will see periods where it is modulating fine, and other periods where it is cycling, but in a controlled manner. It is in tandem with a heatpump, hence the apparently “odd” behaviour (the HEMS decides which to use based on current fuel cost)
Hi Alun
Thanks so much for your reply. I’ve already set d.0 to 15kW in order to reduce wear and tear on the system as it never seems to be able to dump more than 8kW into the house at sensible flow temps.
One thing I’ve noticed is that to get a dT of 12C on the rads, i only have to crack the lockshield valves open 1/8 to 1/4 of a turn. It’s sooo sensitive.
As the heat loss I’ve calculated for the house is approx 7.5kW (outside -3, inside 21) I will never have the boiler in a properly modulating mode, I’m guessing. Basically the boiler is too big! So I need to ensure it cycles in a ‘sensible’ fashion. I’ve currently got d.2 set to 10 mins and d.3 to 15 mins. But the docs are a bit thin on how this affects operation. Do you have any thoughts on suitalbe settings for these?
Cheers
Tim
This is the burner anti-cycling time, when the burner turns off, it should not be turned on again for at least d.2 minutes. However, this is the value for a 20 degC flow temperature. To find out what the actual delay is at the flow temperature you are using you have to consult this table …….
Vaillant anticycle times general.zip (82.4 KB)
The row across the top is the value set in the d.2 register, you then go down the column until you reach the flow temperature, that gives you the actual delay being used. So for a d.2 of 10 and a flow of 60 degC the actual delay is 3.5 mins, so very little effect. If you wanted an actual delay of 10 mins @ 60 degC you would need to put 35 into d.2.
However, looking at the graphs for your system, there are periods where it looks OK. For instance, from 6:30am to 8:00am today it burns for 10 minutes at a time, and then is off for 5 minutes, so 4 cycles an hour. It then goes potty until 9:10am with many short burns, then recovers. So I think your guess of 10 mins looks like a reasonable starting point, so you should put 35 in d.2
d.3 is a measurement and not a setting?
Given the minimum output is 8kW-ish, absolutely correct.
I shouldn’t get too hung up on a dT of 12 degC. In fact, you may be making things worse. By restricting the flow through the boiler the flow temperature will rise more rapidly so you are more likely to trip an over temperature burner shutdown. Assuming the radiators are correctly sized then you are only interested in getting the same dT across all of them. I am running with a typical dT of 5 degC between boiler flow and return.
It would help if you could add boiler flow setpoint to your graph (TargetT). At the moment it is not possible to tell if the cycling is due to setpoint being reached or an over temperature trip.
Hi Alun
I’ve replaced room temp for boiler target temp. Does this help with your analysis?
Sadly my rads are in places quite undersized as they were designed for dT50. In the kitchen our installer put in 2 ‘designer’ rads with 10mm plastic pipe, buried under the tiled floor. Knowing what I now know, this is extremely annoying as the flow through the rads is very slow, and for low temp flow the rads are grossly underpowered. Grrrr!
Re d.1 and d.2 I ‘ve mislead you..sorry 
d.1 is 15, d.2 is 10
d.3 is read-only for a combi, which mine isn’t.
As regards the potty burn from 0800 to 0910, that is where there is call to heat DHW after our showers, and the target temp goes up to 65, to heat the water tank. You can see this on the graph as it’s filled in as ‘white’ rather than ‘yellow’.
If you have any more wisdom I’d be keen to learn as you’ve clearly already been down this rabbit hole!
Cheers
Tim
Yes, can see that is not doing anything crazy. First a general point about the flow rate. From the installation manual the flow for 24kW at a dT of 20 degC is given as 1049l/h. So for the minimum 8kW and dT of 20 degC would be 1049/3 = 349l/h. During the stable cycling periods your dT is 12 degC and flow rate is 612l/h, and 12/20ths of 612 = 367l/h, so about right.
Looking at graphs for Sat 25th Oct ……..
Hi Alun
Thanks for your observations. A few comments below:
From 9:07am to 10:01am it is doing DHW. Flow rate looks fine. The start of this looks reasonable, but after 9:35am it goes unstable. The strange thing is it isn’t even reaching your set point of 65 degC. That looks like it is a safety shutdown due to the flow temperature increasing too rapidly OR it could be that the real set point is still 54 degC OR some d.X value is limiting the max flow temperature.
I discovered that the controls on the boiler actually act as a temp limiter. I’ve whacked both CH and DHW knobs up to 70C and the boiler temp is now making it to 65degC now. Thanks for your pointers here.
From 10:01am to 11:10am it looks fine with stable cycling pattern. Then the flow rate plummets to 144l/h and it enters cycling hell, definitely safety shutdowns as the flow temperature overshoots violently.
This was me. I was trying to get more flow into the distant rads so had shut all the rads off except the ‘index’ kitchen rad. Then gradually opened up the lockshields on all the other rads to get them warm, without starving the kitchen rads of too much. So ignore most of Saturday as I was heavily fiddling.
I would repeat my suggestion of setting d.2 to 35 to try and get more of these long burns. If you could get it down to 2 or 3 cycles per hour that’s probably the best you can do given the gross over-sizing.
I’ve just changed d.2 to 35 mins. I’m not entirely convinced it will have the required effect but let’s see!
Today Mon 27th so far it has looked very stable all day.
no-one was at home to fiddle! The system was just doing its thing.
I’m going to eek the lockshields open a tad more in the hope of increasing flow. I’m still not quite sure if I should be tweaking for good flow or a deltaT of x degC - where x seems to be a variable number depending on who you talk to!
Any more pointers are hugely welcome!
Cheers
Tim
I can see it is having the predicted effect. However, the d.2 delay depends on the flow temperature set-point. The value of 35 was based on a flow set-point of 60-ish, I can see your set-point is 40-ish, so you are getting an actual anti-cycling delay of 20 mins, rather than the predicted 10 mins. But you are getting about 2 cycles an hour, which is good. The question now arises has this affected your comfort with large temperature swings? Your house has a time constant between hours and days, so it can’t see this slow cycling, it just integrates it out. Humans have a much shorter thermal time constant, so may perceive temperature fluctuations. So time to experiment and find a setting for d.2 that is the best compromise between human comfort and reduced cycling.
I can see that with the new slow cycling the pump is shutting off too soon. This means the boiler has no idea of your house load. When you have found your d.2 sweet-spot, I would be inclined to increase your pump run-on time so the pump is running continuously while the boiler is cycling (where it started up at the beginning of the day is as it should look).
You will only get the dT quoted on the data sheet at maximum output, unless the boiler is one of those that tries to control dT by fiddling with pump speed (not many of them). As the boiler reduces power so dT will fall anyway. If your boiler was right sized then it might be worth chasing dT, but with your current system you are more likely to provoke cycling hell by chasing dT.
Ooo, you even got a perfect DHW burn at 11:40am today. Oh, and by the way, d.2 is not active during DHW cycles.
I’m not sure how the boiler knows whether it’s doing DHW or CH. It is a system boiler it was wired for Y-plan, with no input to the boiler for DHW/CH. So the boiler has no clue what it’s doing?
You’re absolutely right, so d.2 should always be active.
