Unvented Expansion Vessel Pre Charge

[Mattg4321]

Hi all

 

Recently had a Joule 250l indirect unvented system fitted. Was told beforehand that incoming main probably needs upgrading. Currently 3/4 inch black MDPE. Around 15mm internal I believe. Plan to do so as soon as funds allow. 
 

When hot water is drawn off after it’s been heated/left for a while then we get a fairly loud gurgling noise from the expansion vessel/cylinder area. 
 

The plumbers been back and I think released some air from the expansion vessel, which seemed to have made it a slightly less aggressive, but not cured. 
 

Im struggling to get them back and just want to get this sorted it it’s something I can do. 
 

I’ve checked pressure on the main and on the balanced cold. 4 bar on the main and 3 bar on balanced cold. This drops to slightly around 0.8 bar with the bath tap fully open. A consequence of the main being undersized I think??

 

Ive checked the pre charge in expansion vessel (after shutting off cold supply and opening a tap to release pressure) and it’s about 1.5 bar. I think it was 3 bar from factory judging by the sticker. 
 

From what I’m reading the pre charge should be 0.2 bar below the dynamic flow pressure of 0.8 bar? Setting it to 0.6 bar seems low though and I don’t want to mess about with something I don’t fully understand without seeking some advice!

 

Thanks in advance 
 

 

[John Carroll]

 

The EV will normally be sized to 10% of the UVC capicity, say 24L for a 250L UVC.

If the precharge pressure is 2.8bar, the final pressure after heating to 60C will be 3.86bar and the EV will contain 5.25L of water, when the dynamic pressure falls to anything below 2.8bar then 5.25L of water will flow out of the EV when HW is drawn off, when HW demand ceases, 1.2L of cold water will flow back into the EV to pressurize it to 3.0bar, after a (if) full reheat, a further 4.05L will enter the EV to give a total of 5.25L.

With a precharge pressure of 1.5bar those numbers will be 4.48bar after heating with 13.1L in the EV most of which will flow back out on HW demand, and 9.0L flowing back in on HW demand ceasing plus a further 4.05L after reheating to give that total of 13.1L.

0.6bar precharge gives, respectively, 5.92bar final pressure with 18.45L in the EV, most of which will flow out on HW demand, 14.4L will flow back in on HW demand ceasing plus a further 4.05L after reheating to give that total of 18.45L.

You can see there is far more water being shunted back and forth the lower the precharge pressure is, the EV is then actling more and more as a accumulator which may or may not be benificial, with a very low precharge of 0.6bar, the final pressure is quite high at 5.92bar, but OK at 4.48bar with a precharge of 1.5bar.

The loud gurgling migh be the combination of a low precharge pressure with a small diameter connecting pipe between the EV and the cold feed.

[Mattg4321]

1 hour ago, John Carroll said:

 

The EV will normally be sized to 10% of the UVC capicity, say 24L for a 250L UVC.

If the precharge pressure is 2.8bar, the final pressure after heating to 60C will be 3.86bar and the EV will contain 5.25L of water, when the dynamic pressure falls to anything below 2.8bar then 5.25L of water will flow out of the EV when HW is drawn off, when HW demand ceases, 1.2L of cold water will flow back into the EV to pressurize it to 3.0bar, after a (if) full reheat, a further 4.05L will enter the EV to give a total of 5.25L.

With a precharge pressure of 1.5bar those numbers will be 4.48bar after heating with 13.1L in the EV most of which will flow back out on HW demand, and 9.0L flowing back in on HW demand ceasing plus a further 4.05L after reheating to give that total of 13.1L.

0.6bar precharge gives, respectively, 5.92bar final pressure with 18.45L in the EV, most of which will flow out on HW demand, 14.4L will flow back in on HW demand ceasing plus a further 4.05L after reheating to give that total of 18.45L.

You can see there is far more water being shunted back and forth the lower the precharge pressure is, the EV is then actling more and more as a accumulator which may or may not be benificial, with a very low precharge of 0.6bar, the final pressure is quite high at 5.92bar, but OK at 4.48bar with a precharge of 1.5bar.

The loud gurgling migh be the combination of a low precharge pressure with a small diameter connecting pipe between the EV and the cold feed.

Thanks for the explanation.

 

The gurgling/vibration noise has got better since the pre charge pressure was dropped to 1.5 bar. It doesn't seem to happen if a basin tap is turned on now. Mostly only the bath. With every other tap its a moderately loud hiss for a while until presumably the EV has emptied. 

 

It seems it's probably something I have to live with until we get the water main upgraded to 25mm blue MDPE? The dynamic flow pressure of 0.8 bar seems to be the root cause here and everything else is a compromise because of this?

[Nickfromwales]

12 hours ago, Mattg4321 said:

Thanks for the explanation.

 

The gurgling/vibration noise has got better since the pre charge pressure was dropped to 1.5 bar. It doesn't seem to happen if a basin tap is turned on now. Mostly only the bath. With every other tap its a moderately loud hiss for a while until presumably the EV has emptied. 

 

It seems it's probably something I have to live with until we get the water main upgraded to 25mm blue MDPE? The dynamic flow pressure of 0.8 bar seems to be the root cause here and everything else is a compromise because of this?

I think what's happening is the expansion vessel is acting like a cold mains accumulator, and the stored energy inside it is reinforcing the flow until it equalises. Basin taps would be better (less noise / less problematic) as the flow rate is slower / lesser, so that makes sense.

 

When you say just had it fitted, is this a retro-fit? 
 

edit: crossed post with @John Carroll’s comment. +1. 

[Mattg4321]

2 minutes ago, Nickfromwales said:

I think what's happening is the expansion vessel is acting like a cold mains accumulator, and the stored energy inside it is reinforcing the flow until it equalises. Basin taps would be better (less noise / less problematic) as the flow rate is slower / lesser, so that makes sense.

 

When you say just had it fitted, is this a retro-fit? 

Yes retro fit into late 60's build with black alkathene main. As per first post I was told that I needed a bigger main, so this isn't a gripe with the plumbers really. More of a question as to whether I can do anything to mitigate the noise in the meantime before I get the main upgraded later this year. Sounds like it may not be wise to drop the EV pre charge below 1 bar, so perhaps it has to be lived with as is now.

 

Plumbers saying 25mm blue MDPE needs to go in. I'm in a fairly large village/small town and water pressure is good (4 bar). Distance from road to stop cock around 15m. Perhaps 25mm is fine, but any benefit to 32mm? I guess it will have to be stepped down both ends to 25mm at meter and 22mm copper in house, so perhaps not really.

[Nickfromwales]

No issue with plumbers, I just wanted to clarify if they've done the install strictly to G3? I see loads of failed setups from the fitter not being 100% clued up about retrofitting an UVC into an existing plumbing arrangement.

 

Is there a 22mm non return valve on the hot outlet, immediately outside the cylinder?

Is there a new second 3 / 3.5bar PRedV installed at the stopcock?

 

Could be down to reverse pressurisation and those potentials shifting from A>B when taps are opened, especially mixer taps, whereas they should all be in equilibrium aka "balanced".

[Nickfromwales]

I've fitted loads of UVC's to existing 15mm cold supplies and they work fine, retrofit rules applied in every single instance.

[Mattg4321]

6 minutes ago, Nickfromwales said:

No issue with plumbers, I just wanted to clarify if they've done the install strictly to G3? I see loads of failed setups from the fitter not being 100% clued up about retrofitting an UVC into an existing plumbing arrangement.

 

Is there a 22mm non return valve on the hot outlet, immediately outside the cylinder?

Is there a new second 3 / 3.5bar PRedV installed at the stopcock?

 

Could be down to reverse pressurisation and those potentials shifting from A>B when taps are opened, especially mixer taps, whereas they should all be in equilibrium aka "balanced".

 

22mm non return valve on hot outlet (at top of cylinder) - not that I can see...

 

No pressure reducing valve at stopcock. Mains pressure at kitchen sink is 4 bar. Everywhere else after combination valve adj to cylinder measured at 3 bar.

 

Not sure what effect those 2 things would have. I've not noticed them on other jobs I've been on either though - not that I make a point of looking. Sure you'll enlighten me! (please)

 

3 minutes ago, Nickfromwales said:

I've fitted loads of UVC's to existing 15mm cold supplies and they work fine, retrofit rules applied in every single instance.

 When bath tap is fully on, flow elsewhere upstairs slows to not much more than a strong trickle!

 

[Nickfromwales]

12 minutes ago, Mattg4321 said:

No pressure reducing valve at stopcock. Mains pressure at kitchen sink is 4 bar. Everywhere else after combination valve adj to cylinder measured at 3 bar.

Ok, then this is a non-compliant install methinks; unless there is NO thermostatic shower / bath mixer taps, and any other mixer taps / mono block taps that don't have separate water pathways to the open ends of the spouts, eg so the hot and cold supplies cannot 'meet' and interfere with each other (aka unbalanced).

 

The PRedV (that is an integral part of the control group valve with the 6bar PRV) that the guys have rightly installed at the UVC gives you a balanced cold outlet for cold taps, and a balanced cold output to go to the cold inlet of the UVC.

 

So I don't quickly assume the worst, lets ask some questions.

 

Was the whole hot and cold pipework altered to have ALL the cold outlets in the house fed ONLY from the control group at the UVC?

Does your existing stopcock pipework now only run, with zero T's off it, direct to the UVC control group?

[Mattg4321]

3 minutes ago, Nickfromwales said:

Ok, then this is a non-compliant install methinks; unless there is NO thermostatic shower / bath mixer taps, and any other mixer taps / mono block taps have separate water pathways to the open ends of the spouts, eg so the hot and cold supplies cannot 'meet' and interfere with each other (aka unbalanced).

 

The PRedV (that is an integral part of the control group valve with the 6bar PRV) that the guys have rightly installed at the UVC gives you a balanced cold outlet for cold taps, and a balanced cold output to go to the cold inlet of the UVC.

 

So I don't quickly assume the worst, lets ask some questions.

 

Was the whole hot and cold pipework altered to have ALL the cold outlets in the house fed ONLY from the control group at the UVC?

Does your existing stopcock pipework now only run, with zero T's off it, direct to the UVC control group?

I do have what I assume to be thermostatic shower. Also a couple of oldish mixer taps. Assume 15 years ish. 
 

Balanced cold feeds everything bar an outside tap and the kitchen tap which is teed off near stop cock

[Nickfromwales]

Just now, Mattg4321 said:

I do have what I assume to be thermostatic shower. Also a couple of oldish mixer taps. Assume 15 years ish. 
 

Balanced cold feeds everything bar an outside tap and the kitchen tap which is teed off near stop cock

Ok, good news, but is the kitchen tap a mixer tap?

 

[Mattg4321]

3 minutes ago, Nickfromwales said:

Ok, good news, but is the kitchen tap a mixer tap?

 

At the moment it’s one where the hot and cold are separated right to the spout. But soon to be changed!!

[Nickfromwales]

3 minutes ago, Mattg4321 said:

At the moment it’s one where the hot and cold are separated right to the spout. But soon to be changed!!

Ok, as is you are compliant.

 

If, however, you do install a new tap (which mixes in the body) then you aren't. That would need the second PRedV at the stopcock and the NRV on the hot outlet of the UVC.

 

Easy going plumbers will tell you it's fine to rely on an added non return valve on the kitchen tap hot feed, but it's just not reliable enough. These start to let-by after a while and then the UVC will get reverse pressurised by the cold pressure at the kitchen sink mixing with the lower pressure hot, and this is particularly problematic when the tap is run on low flow for a decent amount of time.

 

This should have the 2 items, PRedV #2, and a NRV on the UVC hot outlet, to be fully compliant (and to not come back and bite you on the arse in the future). Without these, the UVC expansion vessel will routinely get over pressurised and fail early. 

 

42 minutes ago, Nickfromwales said:

I see loads of failed setups from the fitter not being 100% clued up about retrofitting an UVC into an existing plumbing arrangement.

You chaps are spot on, if they knew the kitchen sink tap was for unbalanced, but they should have assumed that would be changed one day and have gone the extra distance (IMHO). So they get a 9/10 lol, for not getting the kitchen sink fed off the balanced cold.

[John Carroll]

2 hours ago, Mattg4321 said:

Thanks for the explanation.

 

The gurgling/vibration noise has got better since the pre charge pressure was dropped to 1.5 bar. It doesn't seem to happen if a basin tap is turned on now. Mostly only the bath. With every other tap its a moderately loud hiss for a while until presumably the EV has emptied. 

 

It seems it's probably something I have to live with until we get the water main upgraded to 25mm blue MDPE? The dynamic flow pressure of 0.8 bar seems to be the root cause here and everything else is a compromise because of this?

What is the length and diameter of the connecting piping between the EV and the UVC.

[Mattg4321]

7 hours ago, John Carroll said:

What is the length and diameter of the connecting piping between the EV and the UVC.

It’s in 22mm copper and around 800mm

[Mattg4321]

8 hours ago, Nickfromwales said:

Ok, as is you are compliant.

 

If, however, you do install a new tap (which mixes in the body) then you aren't. That would need the second PRedV at the stopcock and the NRV on the hot outlet of the UVC.

 

Easy going plumbers will tell you it's fine to rely on an added non return valve on the kitchen tap hot feed, but it's just not reliable enough. These start to let-by after a while and then the UVC will get reverse pressurised by the cold pressure at the kitchen sink mixing with the lower pressure hot, and this is particularly problematic when the tap is run on low flow for a decent amount of time.

 

This should have the 2 items, PRedV #2, and a NRV on the UVC hot outlet, to be fully compliant (and to not come back and bite you on the arse in the future). Without these, the UVC expansion vessel will routinely get over pressurised and fail early. 

 

You chaps are spot on, if they knew the kitchen sink tap was for unbalanced, but they should have assumed that would be changed one day and have gone the extra distance (IMHO). So they get a 9/10 lol, for not getting the kitchen sink fed off the balanced cold.

Looking like this will be the new tap in the not too distant future. 
 

Assume it means I don’t need to worry as it says it’s protected against backflow with an integrated non return valve?

 

 

[Nickfromwales]

1 hour ago, Mattg4321 said:

Assume it means I don’t need to worry as it says it’s protected against backflow with an integrated non return valve?

👇

10 hours ago, Nickfromwales said:

Easy going plumbers will tell you it's fine to rely on an added non return valve on the kitchen tap hot feed, but it's just not reliable enough. These start to let-by after a while and then the UVC will get reverse pressurised by the cold pressure at the kitchen sink mixing with the lower pressure hot, and this is particularly problematic when the tap is run on low flow for a decent amount of time.

👍

 

Rely on that taps NRV at your peril. I wouldn’t!

 

If this was a clients project there would be no chance I’d leave it like that.

 

My shower mixer had these and I shut the cold inlet off to my combi the other day to work on it and water kept coming out of the hot tap at the sink. When I listened to the shower you could hear water squeezing back through the hot NRV so I then had to turn the temp dial

to max cold to stop it enough for me to do what I needed to, but it goes to show that these factory ones are pretty crap, even with such a good make as you’ve selected.

 

Noise issue you are experiencing is defo the EV acting as an accumulator, and very likely the issue will go when you have a better supply installed.

 

The washing machine etc would be best off the cold mains if practicable, which will give the showers better dynamic performance if the appliances are filling simultaneously for eg.

 

I have a combi so the rule here is if you hear the shower going don’t run taps full blast. Flow just speeds up or slows down with an UVC but with a combi these fluctuations cause havoc (aka entertainment).
 

Anything you can do to improve that is a plus, if doable. 

[Nickfromwales]

19 hours ago, Mattg4321 said:

Ive checked the pre charge in expansion vessel (after shutting off cold supply and opening a tap to release pressure) and it’s about 1.5 bar. I think it was 3 bar from factory judging by the sticker. 

I would pump this back up to 2.5 to stop damage occurring. It was incorrect for this to have been dropped back so far imho. 

[John Carroll]

3 hours ago, Mattg4321 said:

It’s in 22mm copper and around 800mm

 

Maybe we're all barking up the wrong tree!, if these friction loss calculators are to be believed, then 15M of 15mm ID pipe should only lose < 1.0bar at a flowrate of 30LPM, maybe check out the PRV for faulty operation or filter blockage?. You are getting a dP of ~ 2bar at probably around 15/20LPM?.

 

https://www.omnicalculator.com/physics/friction-loss

 

 

 

[Mattg4321]

3 hours ago, Nickfromwales said:

👇

👍

 

Rely on that taps NRV at your peril. I wouldn’t!

 

If this was a clients project there would be no chance I’d leave it like that.

 

My shower mixer had these and I shut the cold inlet off to my combi the other day to work on it and water kept coming out of the hot tap at the sink. When I listened to the shower you could hear water squeezing back through the hot NRV so I then had to turn the temp dial

to max cold to stop it enough for me to do what I needed to, but it goes to show that these factory ones are pretty crap, even with such a good make as you’ve selected.

 

Noise issue you are experiencing is defo the EV acting as an accumulator, and very likely the issue will go when you have a better supply installed.

 

The washing machine etc would be best off the cold mains if practicable, which will give the showers better dynamic performance if the appliances are filling simultaneously for eg.

 

I have a combi so the rule here is if you hear the shower going don’t run taps full blast. Flow just speeds up or slows down with an UVC but with a combi these fluctuations cause havoc (aka entertainment).
 

Anything you can do to improve that is a plus, if doable. 

Thanks. So it seems like I need to get a new main in. 25mm or perhaps even 32mm if there’s any benefit to that? Then fit a 3 bar PRV at stop cock to match the 3 bar PRV at cylinder. Should sort my dynamic pressure problem and potential backfeed problem at new kitchen tap?

 

2 hours ago, John Carroll said:

 

Maybe we're all barking up the wrong tree!, if these friction loss calculators are to be believed, then 15M of 15mm ID pipe should only lose < 1.0bar at a flowrate of 30LPM, maybe check out the PRV for faulty operation or filter blockage?. You are getting a dP of ~ 2bar at probably around 15/20LPM?.

 

https://www.omnicalculator.com/physics/friction-loss

 

 

 

 

I tested dynamic pressure on the main. Ie before the cylinder and PRV. So it can only be a problem on the main or the stop cock (which is almost new and fully open) I think? Guessing that black pipe is just too small. The stop cock is 15mm copper as that’s apparently the size of the black main internally. Then steps up immediately to 22mm copper to cylinder. 

[Nickfromwales]

29 minutes ago, Mattg4321 said:

I tested dynamic pressure on the main. Ie before the cylinder and PRV. So it can only be a problem on the main or the stop cock (which is almost new and fully open) I think? Guessing that black pipe is just too small. The stop cock is 15mm copper as that’s apparently the size of the black main internally. Then steps up immediately to 22mm copper to cylinder. 

None of that should really matter, so it's odd behaviour tbh. Do you have a non return (double check) valve immediately after the stopcock? Defo pump the EV pre-charge back up asap.

 

29 minutes ago, Mattg4321 said:

Thanks. So it seems like I need to get a new main in. 25mm or perhaps even 32mm if there’s any benefit to that? Then fit a 3 bar PRV at stop cock to match the 3 bar PRV at cylinder. Should sort my dynamic pressure problem and potential backfeed problem at new kitchen tap?

That's how I would have left this from the point of commissioning, before leaving the property tbf.

 

Don't forget the NRV on the hot outlet of the UVC, but my advice, with the less than cosmic local network flow / pressure you have, would be to use a 28mm NRV with 22mm internal reducing sets; so there is little to no resistance from the bigger valve body .

[John Carroll]

45 minutes ago, Mattg4321 said:

Thanks. So it seems like I need to get a new main in. 25mm or perhaps even 32mm if there’s any benefit to that? Then fit a 3 bar PRV at stop cock to match the 3 bar PRV at cylinder. Should sort my dynamic pressure problem and potential backfeed problem at new kitchen tap?

 

 

I tested dynamic pressure on the main. Ie before the cylinder and PRV. So it can only be a problem on the main or the stop cock (which is almost new and fully open) I think? Guessing that black pipe is just too small. The stop cock is 15mm copper as that’s apparently the size of the black main internally. Then steps up immediately to 22mm copper to cylinder. 

 

What was the flowrate when you tested the dynamic pressure??, I have a bit of 21mm OD, 11.0mm ID piping lying around here since this house was built 53 years ago, so IF the ID of your pipework is only 11.0mm then 15M at a dP of say 4.0-1.0, 3.0bar will still flow 23LPM, it will still flow 13.5LPM at a dP of only 1.0bar, if your pipework is 15mm ID then it will flow 55LPM at a dP of 3.0bar and 31LPM at a 1.0bar dP., so its important that you know what the flowrate (& dP) was during that dynamic test before you go off doing anything.

[Nickfromwales]

23 minutes ago, John Carroll said:

 

What was the flowrate when you tested the dynamic pressure??, I have a bit of 21mm OD, 11.0mm ID piping lying around here since this house was built 53 years ago, so IF the ID of your pipework is only 11.0mm then 15M at a dP of say 4.0-1.0, 3.0bar will still flow 23LPM, it will still flow 13.5LPM at a dP of only 1.0bar, if your pipework is 15mm ID then it will flow 55LPM at a dP of 3.0bar and 31LPM at a 1.0bar dP., so its important that you know what the flowrate (& dP) was during that dynamic test before you go off doing anything.

Yup. I'll often dig up the mains at the boundary and test there, to discount a damaged or 'furred up' pipe between the network and the stopcock (within the boundary).

 

People have replaced the cold mains and saw little to no improvement.

[Mattg4321]

1 hour ago, John Carroll said:

 

What was the flowrate when you tested the dynamic pressure??, I have a bit of 21mm OD, 11.0mm ID piping lying around here since this house was built 53 years ago, so IF the ID of your pipework is only 11.0mm then 15M at a dP of say 4.0-1.0, 3.0bar will still flow 23LPM, it will still flow 13.5LPM at a dP of only 1.0bar, if your pipework is 15mm ID then it will flow 55LPM at a dP of 3.0bar and 31LPM at a 1.0bar dP., so its important that you know what the flowrate (& dP) was during that dynamic test before you go off doing anything.

 

1 hour ago, Nickfromwales said:

Yup. I'll often dig up the mains at the boundary and test there, to discount a damaged or 'furred up' pipe between the network and the stopcock (within the boundary).

 

People have replaced the cold mains and saw little to no improvement.

I just repeated the test. Pressure measured at the kitchen sink directly off the main. New pipework so won’t be furred up etc 

 

Had the outside tap fully open. Also directly off the main. 
 

static pressure. 3.6 bar

dynamic pressure. 1.4 bar with outside tap wide open

flow rate at outside tap with it wide open. 18 litres per minute 

 

outside tap fed by 15mm copper 

not sure what that tells me!

[Mattg4321]

2 hours ago, Nickfromwales said:

None of that should really matter, so it's odd behaviour tbh. Do you have a non return (double check) valve immediately after the stopcock? Defo pump the EV pre-charge back up asap.

 

That's how I would have left this from the point of commissioning, before leaving the property tbf.

 

Don't forget the NRV on the hot outlet of the UVC, but my advice, with the less than cosmic local network flow / pressure you have, would be to use a 28mm NRV with 22mm internal reducing sets; so there is little to no resistance from the bigger valve body .

Just checking the EV instructions. I think it says 65% of max pressure. So 65% of 3 bar. Equals 1.95 bar

 

However there is a pre charge acceptance factor table. But I don’t really know what it relates to at a glance through instructions.