Gurgling and air in central heating

[jim843]

Hello,

Around 2 years ago we had a very old, leaky Baxi combi replaced with a newer Baxi.

For background I live in a 3 storey terrace. First floor has 3 rads, second floor has 4 rads and third floor has 2 rads.

We noticed problems with noise and air in the system. One of the rads in the loft made gurgling noises. If the rad was bled, air came out. Also, at this time the pressure gauge was all over the place. Up to 3, back down to 1 and every number in between. The gauge didn’t appear to drop, though.

The installer wasn’t helpful over the phone so I ended up finding a different engineer.

Turns out the installer of the boiler didn’t do a particularly thorough job. He didn’t clean the old system, I found leaks on some of the nuts beneath the boiler, he didn’t fit a soak away outside so water drips out over our path (great in winter when it’s freezing).

He engineer found the expansion vessel was flat. He pumped it up, tightened the pipes, advised on a soak away. He told me to bleed all the rads individually for a few mins, top up the system and add more inhibitor. He also suggested power flushing the system. I did all this and also rented a machine and flushed the system myself.

I did all this but the problems remained. I ended up contacting Baxi who came out and replaced the expansion vessel.

This didn’t solve anything. More gurgling, more air.

We ended up getting the boiler serviced again (this is maybe a year after install) and I mentioned the problems to the engineer. He asked the usual questions (is there inhibitor, have you bled the system, etc) and mentioned we had an old radiator in the bathroom. He thought this might be the source of air.

We were planning on getting the bathroom refitted so I decided to wait and see if that solved any issues.

We got the bathroom refitted. Now, the gurgling and air does not go to the top of the house but is in the towel rail in the bathroom. I can bleed the rad once a day and there’s maybe a second or two of air and then water. The pressure does not drop on the boiler.

I contacted Baxi again who came out and replaced the expansion vessel with a bigger one. The engineer also suggested fitting an automatic bleed valve on the chrome towel rail because in his experience chrome towel rails suck in air. I bought one two valves (one for spare) and fitted it.

I noticed an odd thing while fitting the automatic bleed valve. I turned off the two rad valves, released the pressure. Once the air left the rad I could put my finger in the top of the rad and feel there was no water there at all. I’d expected that the water would be just below the top of the rad?

But you’ve guessed it — still gurgling and air in the system! The automatic bleed valve doesn’t appear to vent automatically, not sure why. Could be faulty. I can swap it over with the spare but haven’t gotten around to it yet.

The only other thing I noticed with the system is that if the system is off in the summer then there is minimal air in the rad. I assume because the pump isn't forcing water around the system?

If you got this far, thanks! I’d love to solve this issue and have a system that works! I’m not opposed to ripping out pipework and starting again (even though I’m nearing the end of a long renovation!). Ideally there are some simpler steps I can take to diagnose?

Any advice?

 

[jim843]

Was the new E.vessel installed internally in the boiler or is it external where you can see it!.

Pretty sure it’s internal. The engineer took the cover off. There’s nothing outside the boiler as far as I know.

 

[John.g]

In the attachment is that a PRV on the filling system??.

It's also possible I suppose but very unlikely as I said above that the pump is pulling in a tiny amount of air through that leaking NRV each time it starts especially if the E.vessel is on the discharge side of the pump, the closer to the boiler the greater the chance, if the boiler is under warranty then that should be covered? its the only part that has direct communication to the atmosphere via the leak even though pressurized.

 

[jim843]

In the attachment is that a PRV on the filling system??.

It's also possible I suppose but very unlikely as I said above that the pump is pulling in a tiny amount of air through that leaking NRV each time it starts especially if the E.vessel is on the discharge side of the pump, the closer to the boiler the greater the chance, if the boiler is under warranty then that should be covered? its the only part that has direct communication to the atmosphere via the leak even though pressurized.

Maybe I’m mixing this up: the leak is on the valve that has an arrow on it.

The arrow you have drawn on the photo: do you mean a pressure release valve? I don’t know what that black thing is.

 

[jim843]

In the attachment is that a PRV on the filling system??.

It's also possible I suppose but very unlikely as I said above that the pump is pulling in a tiny amount of air through that leaking NRV each time it starts especially if the E.vessel is on the discharge side of the pump, the closer to the boiler the greater the chance, if the boiler is under warranty then that should be covered? its the only part that has direct communication to the atmosphere via the leak even though pressurized.

This is where the leak is (see attached).

Do you think it’s worth replacing that NRV?

 

[John.g]

Yes, I would replace it as every other possibility seems to have been reasonably exhausted. Maybe its under warranty?.

The valve thats leaking is a non return valve (NRV). the PRV I'm referring to is a pressure reducing valve called a PRV (as is the safety valve also called a PRV, a bit confusing) the pressure reducing valve above is/should be connected dirrectly to the filling hose?.

Forget my reference to that PRV, thats the isolating valve for the filling hose.

 

[John.g]

As a temporary measure you could attach another filling hose which has isolating valves at both ends, connect one end to the NRV and shut it.

 

[jim843]

Yes, I would replace it as every other possibility seems to have been reasonably exhausted. Maybe its under warranty?.

The valve thats leaking is a non return valve (NRV). the PRV I'm referring to is a pressure reducing valve called a PRV (as is the safety valve also called a PRV, a bit confusing) the pressure reducing valve above is/should be connected dirrectly to the filling hose?.

Forget my reference to that PRV, thats the isolating valve for the filling hose.

I seem to remember from a couple years back a Baxi engineer saying that this filling loop is old and should have been replaced when the boiler was replaced. Maybe I’m misremembering, though.

You’d think the boiler would be under warranty but the installer has moved and I’ve no idea where they live now. Considering the mess they made - leaking joints under the boiler, not fitting the soak away - I should probably get someone else in.

 

[jim843]

As a temporary measure you could attach another filling hose which has isolating valves at both ends, connect one end to the NRV and shut it.

Ok, I’ll have a look and see what screwfix have tomorrow.

 

[jim843]

As a temporary measure you could attach another filling hose which has isolating valves at both ends, connect one end to the NRV and shut it.

Something like R24 Filling Loop - https://www.NoLinkingToThis/p/r24-filling-loop/83905 with some pipe and compression fittings to connect to that new loop?

 

[jim843]

As a temporary measure you could attach another filling hose which has isolating valves at both ends, connect one end to the NRV and shut it.

Another thought: the isolation valve with the handle before the filling loop would stop the leak, if any, right?

Or are you thinking that the leaking NRV is somehow pulling air into the system even though the filling loop isolation valve is closed?

 

[John.g]

Yes, maybe through the leaking NRV but very unlikely.

The most likely place for air ingress one might think would be at at attic rads even though these should still have ample positive pressure even if boiler pressure was =<1.5bar.
If you vent them with pump running does water definitely flow out after a burst of air? and does the same apply to all the other rads?

 

[jim843]

Yes, maybe through the leaking NRV but very unlikely.

The most likely place for air ingress one might think would be at at attic rads even though these should still have ample positive pressure even if boiler pressure was =<1.5bar.
If you vent them with pump running does water definitely flow out after a burst of air? and does the same apply to all the other rads?

The attic rads never have any air in them. They only had air in them before the bathroom was fitted. After the bathroom was fitted, air appeared in the bathroom towel rail instead.

I can always cap off the upstairs rads on the second floor and run the system for a bit to see if the air stops?

If the attic rads are pulling in air, why would it end up one floor down?

 

[John.g]

Its a mystery to me why any air is getting in!, does the towel rad vent water??.

 

[jim843]

Its a mystery to me why any air is getting in!, does the towel rad vent water??.

Yep, it vents water.

There’s some air. Then water.

I can repeat this every day.

 

[jim843]

Its a mystery to me why any air is getting in!, does the towel rad vent water??.

I re-read your previous comment: did you say vent with the pump running? I always vent when the pump is off.

Also, air only appears in the towel rail when the heating system has been on. In summer there’s no air in the towel rail. This might be expected, I don’t know.

 

[John.g]

No, vent with pump off but when water is circulating in any pressurized/sealed system then there should always be water pressure at the vent when opened, if no water then ensure that air is actually venting and not being drawn in.

I have two rads in a upstairs attic conversion that always run with a slight vacuum (open vented system) but it has never caused any problems because there are no leaks anywhere associated with them.

 

[jim843]

No, vent with pump off but when water is circulating in any pressurized/sealed system then there should always be water pressure at the vent when opened, if no water then ensure that air is actually venting and not being drawn in.

I have two rads in a upstairs attic conversion that always run with a slight vacuum (open vented system) but it has never caused any problems because there are no leaks anywhere associated with them.

There’s always pressure. The air is always forced out and I get water.

 

[John.g]

OK, Thanks, stumped for sure, can you please post the exact model of the Baxi Combi you have.

Also, what is boiler flow temperature on CH.

 

[jim843]

OK, Thanks, stumped for sure, can you please post the exact model of the Baxi Combi you have.

Also, what is boiler flow temperature on CH.

Sure, it’s a Baxi Duo-tec Combi 33.

I think it’s anything from about 60 to 80. Will look more closely.

 

[John.g]

At least one small query is answered in the schematic, the E.Vessel is on the flow side, the reason why the pump pressure drops on start up.

 

[SJB060685]

If that schematic is correct then perhaps we have the answer. The pump is pumping towards the point of no pressure change and therefore the system is under negative pressure. Any weak joints will be drawing air in, alternatively the the system being under negative pressure means you will pull any dissolved gases out of solution.

 

[John.g]

The point of no pressure change is where the EV connects into the system and is the EV pressure itself, with the circ pump off then this pressure will be exactly the same throughout the circuit, after allowing for elevation. With the circ pump running, while the pressure where the EV connects to the system will still be exactly the same then its possible to get negative pressure(s) somewhere but only due to pipe friction losses due to blockages?.
Maybe all combis have their EVs connected in likewise, someone might care to watch their own boiler pressure gauges on start up and note whether it rises or falls with pump start and also its magnitude, the one above certainly moves more than I would have expected, even though theoretically possibly correct, possibly pointing to restriction somewhere?

If the system isn't ingesting air then all this air must be trapped since initial filling but as its being slowly vented then the boiler pressure should gradually fall and require top up but it isn't. Perhaps drain the whole system again and fill very slowly stopping to vent from the bottom up.

The only other thing maybe worth checking out is the PHE, if it had a pinhole maybe possible to keep the system nicely topped up from the mains and if its at or near the system pressure then no problem with boiler PRV lifting. but plenty of oxygenated air gaining access.
Suggest shutting off the mains feed at the boiler and open a hot tap and leave open for a few hours with CH in service and see if any change in behaviour, presume the diverter valve isolates the primary water from the PHE but something may still show up.

A interesting riddle alright.

 

[SJB060685]

If the pump is pumping towards the vessel then the entire system will be under a negative pressure, from the pump suction side right the way back to the vessel.

 

[John.g]

Don't see it that way SJB but could be wrong, pages 36/37 of the attached explains even though they use very low pressures to demonstrate how a negative pressure may occur.

https://www.caleffi.com/sites/default/files/coll_attach_file/idronics_16_na_0.pdf

 

[SJB060685]

Yes, so my above comment wasn't worded correctly. The pump will maintain a differential pressure though. Therefore if it's only able to increase pressure slightly on the discharge the rest will be seen as a negative pressure on the suction side.

 

[jim843]

The point of no pressure change is where the EV connects into the system and is the EV pressure itself, with the circ pump off then this pressure will be exactly the same throughout the circuit, after allowing for elevation. With the circ pump running, while the pressure where the EV connects to the system will still be exactly the same then its possible to get negative pressure(s) somewhere but only due to pipe friction losses due to blockages?.
Maybe all combis have their EVs connected in likewise, someone might care to watch their own boiler pressure gauges on start up and note whether it rises or falls with pump start and also its magnitude, the one above certainly moves more than I would have expected, even though theoretically possibly correct, possibly pointing to restriction somewhere?

If the system isn't ingesting air then all this air must be trapped since initial filling but as its being slowly vented then the boiler pressure should gradually fall and require top up but it isn't. Perhaps drain the whole system again and fill very slowly stopping to vent from the bottom up.

The only other thing maybe worth checking out is the PHE, if it had a pinhole maybe possible to keep the system nicely topped up from the mains and if its at or near the system pressure then no problem with boiler PRV lifting. but plenty of oxygenated air gaining access.
Suggest shutting off the mains feed at the boiler and open a hot tap and leave open for a few hours with CH in service and see if any change in behaviour, presume the diverter valve isolates the primary water from the PHE but something may still show up.

A interesting riddle alright.

If I understand: the EV can cause negative pressure in the system. So a weak point/pin hole leak could be anywhere?

Your suggestion is:

1. switch off the main feed (the cold water?)
2. switch CH to service (not sure how to do this? Just switch to hot water only?)
3. Run hot tap in the kitchen for a couple hours
4. Watch for pressure drop/change?

 

[jim843]

@SJB060685 and @John.g...

I'm off next week so will fit an isolation valve to the leaky NRV. I'll also try the steps suggested above.

Sounds like I'm getting closer to the dreaded "replace all the pipework in the house"?

Supposed to be getting carpets fitted in 3 weeks. The last in the house after a 7 year renovation in my spare time. Oh well, better know now before getting the carpet fitters in...

 

[John.g]

If I understand: the EV can cause negative pressure in the system. So a weak point/pin hole leak could be anywhere?

Your suggestion is:

1. switch off the main feed (the cold water?)
2. switch CH to service (not sure how to do this? Just switch to hot water only?)
3. Run hot tap in the kitchen for a couple hours
4. Watch for pressure drop/change?

I don't think the EV is causing negative pressure Jim.

Re 1 to 4 above .

Shut either valve 13 in schematic post 80, or just shut the main stopcock, either will cut off the cold water supply to the combi side of the boiler, open any hot tap, (hot) water flow should stop.
Ensure boiler on CH say like now, just watch the pressure gauge for few hours for any change.
When complete shut hot tap and restore the cold water.

Just trying to establish if there is a leak between the mains and the boiler "water" via the HW heat exchanger. (PHE).

 

[jim843]

I don't think the EV is causing negative pressure Jim.

Re 1 to 4 above .

Shut either valve 13 in schematic post 80, or just shut the main stopcock, either will cut off the cold water supply to the combi side of the boiler, open any hot tap, (hot) water flow should stop.
Ensure boiler on CH say like now, just watch the pressure gauge for few hours for any change.
When complete shut hot tap and restore the cold water.

Just trying to establish if there is a leak between the mains and the boiler "water" via the HW heat exchanger. (PHE).

Ok, gotcha. Thanks.

I can do this tomorrow (probably -- weather dependent!).

Do you want me to run the CH as normal (Tado switches various rooms on/off during the day)?

 

[John.g]

Just as normal.