Boiler - Navien NCB300 Cycling

[machine-bit]

I have managed to get SAT behaving quite nicely. But I am still getting cycling with my boiler. I have got most attributes stable but often get the flame health stating problem.

I assumed that the boiler would start at a low temp and as the return temperature started to rise the boiler would increase in temperature whilst the system was still calling for heat to stop it cycling?

This is a couple of screen prints. I don’t think the flame health stating problem is helping but struggling on how to troubleshoot.

[sergeantd83]

Can you please provide a boiler flow water temperature vs return water temperature vs control Setpoint 1 hour graph? It will help to identify the issue.

[machine-bit]

@sergeantd83 thanks for your help! I owe you a beer, this is the graph overnight, note I am using the entities from the opentherm and the central boiler from Versitile for the calling for heat. Does this help? I dont know why its sending a setpoint of 10.

It should be noted that I am running "Dynamic Minimum Setpoint (Experimental)" plus "Minimum Setpoint Controller Version" of Boiler Temperature,

[machine-bit]

Hi, Ihave got it working slightly better, still someting is making teh flame extinguish every 30 mins

think its something to do with it recieving a max modulation for 50

Other things going on Relative Modulation Sync "Problem" - constantly
I think the flame drops correspond to Flame Status = STUCK_ON to IDLE to IDLE_OK and I think corresponds to Boiler Status changing to OVERSHOOT_HANDLING

[sergeantd83]

Ok, so the GEMINOX option did the trick and is working? If yes then it is huge.

[machine-bit]

I think it helped, I want to troubleshoot these issues too. If I enabled debug is the extra logging (assuming I enable debug) just put into core logs?? My aim is to get the boiler to run long and low fro as much as possible. I have just disabled SAT as something is ramping the max modulation upto 50% and then back down again, this caused the t_tset to drop to 10. Maybe this shows it better?

[sergeantd83]

You can change the duration of each cycle from 4x to 3x and see how it behaves. Probably this causes the issue.

[machine-bit]

Cheers I have done that, I still think thingthing else is at play with the boiler as its still modulating to zero and causing the boiler to cycle even without the SAT running, I am getting some timeouts in the opentherm logs so I might try and replace the comms cable for a CAT6 piece of cable. Wonder if teh time outs cause a wobble

[machine-bit]

It was going in the right direction, but then this happened

Any ideas? Also should I be settign the heat mode to off when nothing will be calling for heat? It seems to fire up a few times an hour (maybe th cycyling routine?)

I have also got some orphaned entities whowing up rooms (from verstitle, now removed) any tips on how tio remove? I tried a couple of things last night including spook and was even gpoing to manually remove from core.entity_registry json

Thanks

[sergeantd83]

This second graph is the perfect example of how Low-Load control works.

Also should I be settign the heat mode to off when nothing will be calling for heat?

No, because the PID algorithm will be confused.

I have also got some orphaned entities whowing up rooms (from verstitle, now removed) any tips on how tio remove?

You need to reconfigure SAT and then remove these orphaned entities from the Areas section.

[machine-bit]

Argh ok, so that low load is a good thing as it’s keeping it topped up and better than letting it cool down and take a lot longer to get to temperature again? If I left it like that 24/7 assuming the UFH heating was calling for heat 24/7 is that not a bad thing for the boiler?

Thanks I appreciate you taking the effort. Maybe I can write you a numptys guide to SAT in return? Haha

[sergeantd83]

so that low load is a good thing as it’s keeping it topped up and better than letting it cool down and take a lot longer to get to temperature again?

Yes — especially with underfloor heating (UFH).

UFH systems have very specific physical characteristics:

1. Large thermal mass (the slab stores a lot of energy)
2. Very slow response time
3. Designed for steady, low power input, not on/off bursts

If you allow the slab to cool down and then reheat it:

1. It takes hours to recover
2. The controller is forced to request higher flow temperatures
3. Boiler modulation increases sharply

Overshoot becomes much more likely.

Low-load control does the opposite:
It continuously supplies only the heat that the building is losing, keeping the slab in thermal equilibrium instead of cycling it.

This is not “keeping it hot for no reason” — it’s preventing unnecessary thermal swings.

If I left it like that 24/7 assuming the UFH was calling for heat 24/7, is that not bad for the boiler?

Of course not, because the boiler isn’t stressed at all. It operates smoothly most of the times.

[machine-bit]

Hi, I am still trucking with this obsession haha, I assumed that having the boiler fire up for 16 mins and then 4 mins off wouldnt be great for the boiler. The return isnt peaking too much so surely no need to cycle?

Also I have tried all night but these room climate seem to be sticking around nomatter what entities I delete / files I clear up

[machine-bit]

Hi, I am still trucking with this obsession haha, I assumed that having the boiler fire up for 16 mins and then 4 mins off wouldnt be great for the boiler. The return isnt peaking too much so surely no need to cycle?

Also I have tried all night but these room climate seem to be sticking around nomatter what entities I delete / files I clear up

I sorted this, only way I could remove them was to delete and call the device something different