Bubbles appearing regular as clockwork but only after boiler turns gas off ... why?

[Kesterlester]

We have an open vented S-plan system with an old (but ultimately quite reliable) Ideal Classic NF50 gas boiler. This is one of those older systems where a small controller is in charge of the two zone valves, the pump and the power to the boiler. When one (or both) valves are open and the pump is on, the boiler is powered and makes its own decisions (based on its thermostat) as to when it should burn gas. All radiators have TRVs. There is a bypass valve that should allow water to circulate when all TRVs are closed. [Whether the bypass valve works I don't know.]

On a typical wintery day, with the CH on and the pump running all the time, the normal boiler cycling process sees the boiler burn gas for about 1 minute, then turn the gas off for about 5 minutes, with this cycle repeating continuously.

While that cycling rate is high, it's happily continued doing this (when required) since 1992, so I have no desire to change the above, or upgrade the boiler.

However, just recently, I have noticed lots of fine/small bubble sounds appearing in the pump about 10 seconds after the gas turns off in each cycle. The pump bubble noise lasts about 20 seconds. After they stop, these bubbling sounds don't reappear until the same time in the next cycle. For example, if the boiler is prevented from firing (by turning it off at its isolator) the bubbles never reappear, even though the pump keeps circulating.

Ten seconds is about the time that it would take the pump to get water from the boiler to the pump (and the pump is the first thing that the water from the boiler reaches), so it seems likely to me that the source of the bubbles is somewhere inside the boiler, and that --- for some reason I can't figure out -- the bubbles only enter at the point that the boiler decides to turn gas turns off. Immediately after the pump is the vortex thing below the expansion pipe to the attic -- so the bubbles appear to be being successfully rid from the system either in the pump spindle or the vortex thing, as they don't come back until the next equivalent point in the cycle.

Can any of you suggest a reason why the bubbles would only appear at this point in the cycle?

That's my basic question.

Aside:

I did replace a small towel radiator three days ago, and I'd naturally expect to get (and did get) a small amount of air entering the system as a result. [Not much entered, as there were plenty of isolating valves ... and in any case the radiator was small]. However, I don't think the bubbles I am talking about in my question are related to the radiator change as they only correlate with the turning off of the gas at the boiler, and do not correlate with any other general pump operation. I therefore suspect that the current bubbling thing has been going longer, but I only noticed it after the above radiator fitting as I was paying more attention to the system bedding in. I should add that there are no big bubbles anywhere collecting in any radiators. The sounds represent (at least in my imagination) thousands of small sand-grain-to-lentil-sized bubbles, not the big blobs of air.

 

[MJN]

Just thinking aloud until the pros turn up, whilst you've correlated the bubbling with the gas valve turning off that is also presumably the point of the boiler stat temperature being reached i.e. the hottest it gets. What temperature is this? Might it be boiling (perhaps just locally at the heating junction) and thus evaporating the water vapour which will then condense back as it cools (after the pump)? (Vapour or perhaps dissolved air from the new water but I would expect that to eventually stop)

 

[Kesterlester]

Dear @MJN , Yes, you are correct that the gas going off does indeed correspond to the hottest part of the cycle, as the gas turning off is controlled by the boiler detecting its cutout temperature. I don't know what that temperature is exactly in centigrade or Fahrenheit, as I've never measured it, however I do know that the relevant knob on the front of the boiler can vary that cutoff point, and that knob/function does work. At the moment it's at a mid to low setting (3 out of 5 or 6 or whatever the scale is). If I decrease that a bit, then the cutout temperature is indeed lower, and the bubbling is a bit less pronounced, but still starts 10 seconds after the gas cutout and stops 30 seconds after the cutout.

I think there could well be something in your idea of gasses coming out of the water during small pockets of boiling inside the heat exchanger, as there is (and for many years has been) a small amount of kettling going on in there. The boiling of the water itself cannot be the cause, as a steam bubble will instantly disappear (well, in much less than a second) and could never reach the pump 6 meters away. But, as you righty point out, if oxygen or nitrogen or some other gas can be brought out of the water into bubbles by the boiling process, then those bubbles would not disappear as quickly as steam would, and they could get to the pump. And, as you rightly point out, (1)the fresh water that the system gained when the towel radiator was added would have introduced some fresh dissolved gasses into the system, and (2) that should gradually disappear with time ...

... but ....

... the one thing that (for me at least) seems not to fit the above hypothesis is that the low level kettling noises appear already for many seconds before the boiler thermal cutout is reached, and during all that time there are no bubbling noises at the pump. Indeed, as soon as the gas turns off, the pump is pretty much silent for a good 10 secs (other than the sound of its motor) .... before gradually the bubbles arrive in a solid stream. So the tricky bit of physics is to explain why the 'early' boiling/kettling doesn't lead to bubbles which reach the pump, yet something which can happen after the gas goes off can lead to bubbles reaching the pump.

At the moment by best guess is that a joint opens somewhere inside the boiler when some part contracts on cooling, allowing some bubbles to be brought in by a venturi effect. But I find even this implausible as (a) I can't find any leaks in the boiler (though admittedly I can't see behind or on top of the heat exchanger), and (b) I struggle to imagine the venturi effect really being strong enough under a 4m head of water.

Therefore it would be good if your suggestion were to turn out to be true ... if someone could explain why the air bubbles would choose to come air come out only (or 99% of them) after the gas turns off.

 

[MJN]

Hmm... You've got me there! It's certainly a head scratcher (to me at least - to someone else more familiar it might be a straightforward one to explain). Will be following with interest!

 

[iateyoubutler]

I`m glad that you asked this question because I used to have this exact problem with my Vaillant ecotec heat only boiler. When the central heating was on all was fine, if I had the hot water only on and the boiler was firing there would be huge amounts of gas/steam/air (whatever) going up through the pump, to the point where it sounded like it was trying to pump soap suds, and terrible gurgling from the hot water cylinder. If I turned the boiler off and just ran the pump, there would be no noise whatsoever, so it had to be a boiler issue.......

I never did get to the bottom of if, but I had one of those air separators fitted in before the pump - those things are brilliant, and no more issues since.

It`s strange, because the boiler that it replaced (a 1970`s Ideal) never had this issue, it worked perfectly. Along came the Vaillant and cue this issue that you have.

I`m watching this thread with interest!!

 

[gasbanni]

It's hard to come up with a solution without beingu there and seeing it.

My thoughts are it's an old system and probably needs a good flushing out. When the heating's been on a while cup your hand around the bottom of a radiator in the middle. If it fails to heat up and the rest is hot it's sludged up ( ferrous oxide). Plus the boiler can be sludge up too.

When the heating is on listen carefully at the boiler can you hear a kettling/ slight bubbling noise ?
Or does the boiler bang a little ? Another sign that your heating is sludged up.

If the heat exchanger has sludge in the bottom it prevents good contact between the hot heat exchanger and the water so it can get too hot and bubbles form.
Anyway that's my pennies worth.

 

[Kesterlester]

I insert the attached solubility graphs, (see foot of post) as I have to admit (ahem!) that I didn't know how very differently the solubility curves of gasses behave compared to those of salts/sugars/ionic-solids.

From my school chemistry I had only remembered solubility curve which rose with temperature. This is true for sugar (i.e. it's easier to dissolve sugar in hot tea than in cold tea), and most solids, but the reverse is true for gasses, it seems. I'd forgotten this latter point from school!

Therefore, indeed, the fresh (and very cold wintery) water supplied when fitting the radiator will have likely brought with it more dissolved gas than I had previously imagined it would, and indeed that has to come out as the water is heated and circulates. And as @MJN said, more has to come out at the peak of the temperature curve, which is close to when I hear it. I am therefore happy/confident that an internal (i.e. non leak-based) source for the dissolved gas is now identified: it was the radiator addition, contrary to my initial presumption.

Now all that remains for me to understand is the mechanism that prevents release prior to cutout.

I am prepared to believe that some sludge dynamics could potentially contribute as @gasbanni suggests. I can't check the radiators right now (in the spot he mentions) as its now all off for the night, but I can do so later, and may ask a plumber to come round and desludgeify. I have always tended (is this unfair of me?) to assume that most people with just a powerflush machine are really just selling snake oil ....... and that to do anything properly the heat exchanger should come out and be flushed independently after a number of fillings with some noxious chemicals. If so, this is something I hope someone in MJPotts of Cambridge might do.

Before signing off for the night, my last thought for the mechanism is that when the gas is cut, so is the forced air flow to/from the boiler. Perhaps this means that in the last second after cutoff the hot combustion gasses heat parts of the exchanger that are normally being kept cool by incoming air. If those are parts of the exchanger harbouring sludge or something that's good at trapping bubbles, perhaps those sludge heaps get a final push to liberate these bubbles in the resulting atypical 'heat bath' associated with the loss of pumped air.

 

[gasbanni]

Very good !