Noisy return

[John Carroll]

On 06/11/2024 at 11:03, Little Clanger said:

Recently, I've had to change our heating system. Previously, we had an indirect hot water cylinder, on the ground floor, near the Stanley solid fuel range that heated the water in the cylinder.

 

I've installed a new Economy 7 hot water cylinder upstairs for the DHW, and replaced the old, downstairs one with a 250l Economy 7 cylinder, to act as a thermal store. for the central heating circuit. As far as possible, I've left the existing pipework in place. This included the pump on the return, as fitted by the plumber about 40 years ago (he said something about putting it on the return to avoid 'over topping', which I assumed meant the risk of pumping the flow up into the F&E tank in the loft, via the swan neck expansion pipe).

 

It all seems to work OK with the exception of noise building up on the return. When the pump first turns on, its so quiet that I have to check it's running. After about 15 minutes, the return pipe near the pump starts to get noisy, sounding like - not surprisingly - gushing water. After a while the noise gets annoyingly loud. I didn't experience this with the previous set up. I've attached a video to give some idea.

 

So basically, all that is different with the new system is that now the water in the radiator circuit is being drawn from the thermal store, rather than from the Stanley - cylinder coil circuit. And that I've replaced the ancient Grundfoss pump with a new Trident one (I tried changing the pump back again, but it made no difference).

 

I had a mad theory that drawing the water form the thermal store is too demanding, and that the system is actually drawing air in from the F&E swan neck, as that is easier. Once I shake off this fluey cold, I'll get up in the loft and check. But in the meantime, has anybody any idea what is happening? Incidentally, if my mad theory is true, can I extend the swan neck down below the water level of the the F&E tank, with a non-return valve?

 

Thanks

 

 

 

 

With the pump running, hold a "glass" of water with the end of the vent immersed in it and see does the level start falling.

[Little Clanger]

4 hours ago, John Carroll said:

 

With the pump running, hold a "glass" of water with the end of the vent immersed in it and see does the level start falling.

Brilliant idea, thanks. Presumably a falling level in the glass means it's drawing air in? I did try stuffing a cork in the vent - temporarily - but that made no difference.

Do I even need a vent now I don't have a boiler? Presumably any expansion in volume as the water heats up would be accommodated by the F&E pipe, since even if a thermostat goes, they have a safety cut-out?

[John Carroll]

If you remove the vent then the now combined feed & vent must be 22mm with no isol. valve.

What is the height  roughly from the F&E cistern water level to the pump and what is the pump set to, speed 1,2or3?. If that height is greater than the pump head, then the pump suction could possibly be running at a negative head and eventually pulling air in, for example if the distance is say 3M and the pump head is 4M, then the pump might be running with a negative suction head of 1M.

Anyway try the glass test and give a rough idea of the height and pump speed.

Did you have a gas or oil fired boiler previous to the change?

Edited Tuesday at 15:09 by John Carroll

[Little Clanger]

OK, John, so the height from the pump to the water level on the F&E tank is 3.7m. By 'the pump head', I assume you mean the suction pressure of the pump. The blurb for the pump (a Trident) only says that the maximum head is 6m (presumably maximum pressure when pumping on the flow). The pump is set on the lowest speed, 1, although actual rpm aren't given, and I've a feeling it runs a bit faster than the old Grundfos. It uses a bit more power.

With the glass of water test, the level stayed the same, but when the pump first started up - before I'd got the glass under the vent - water was pumped out of the vent a couple of times.

Previously, we had a big Stanley solid fuel range, burning wood.

Thanks 

John

[Little Clanger]

Should have been more specific. The old Grundfos is a UPS 15-50. The Trident is a WRS 25/60-130, listed as a replacement for the 15-50.

[John Carroll]

If, which it is, the pump is pumping over on start up, then its more than likely that its doing the same on stopping so will eventually draw in air from the vent, your pump (I have the pump curves) will only develop a head of 1M on speed1, a bit surprising that that head is sufficient to heat all your rads, but shouldn't in itself cause that pump over except it starts up at full speed (6M) and then ramps down

 which I wouldn't expect it to, you said it uses a bit more power than the Grundfos, does it display this power (W), what is the power wherever you are getting it.

The pump over often points to a blocked or partially blocked cold feed where it joins the flow pipe, in one of your posts above you show the cold feed & the vent where they are teed into the flow pipe, is the flow from left to right or right to left?.

What was the pump power and where did you get that from?.

[Little Clanger]

In that photo, the flow is from right to left, so hits the 22mm vent first (although of course the pump is on the return, so drawing through the flow rather than pumping).

I gleaned the power data from the labels on the pump. Not directly comparable as the Trident is derived from 220v, the Grundfos 240v. The Trident draws more current (0.25 amps compared with 0.17), which at 240v gives 60W compared with 40.

[John Carroll]

The only reason I can give for noisy operation now vs with the stove is that the stove probably had a much higher resistance to flow because probably a build up of sludge etc over the years compared with the new cylinder resulting in the pump over on start up & (probably) shut down, is there any possibility of raising the vent pipe higher before it turns down?. I am very surprised still at this pump over with a pump developing only 1.0M head, normal pump head would be ~ 3/3.5M, like mine at 3.6M, same set up as yours, pumping into the boiler return but with a "combined" vent & cold feed, they are combined at the F&E cistern. (but I still retained the vent), I have a oil fired boiler with little or no resistance to flow.

 

 

Edited Tuesday at 21:59 by John Carroll

[Little Clanger]

Thanks for all your help, John. The Stanley had a huge boiler, so that could explain the difference. A higher vent would be difficult, but not impossible, so I'll give it a try.

Thanks again 

John

[marshian]

7 hours ago, John Carroll said:

If you remove the vent then the now combined feed & vent must be 22mm with no isol. valve.

What is the height  roughly from the F&E cistern water level to the pump and what is the pump set to, speed 1,2or3?. If that height is greater than the pump head, then the pump suction could possibly be running at a negative head and eventually pulling air in, for example if the distance is say 3M and the pump head is 4M, then the pump might be running with a negative suction head of 1M.

Anyway try the glass test and give a rough idea of the height and pump speed.

Did you have a gas or oil fired boiler previous to the change?

 

Sorry to thread jack a smidge but thank you for that information - I've got a pump head of 3.0 m but the actual height of the F&E tank water level to the pump is closer to 2.0 m

 

Might explain why I get occasional issues with air in the system with no apparent signs of leaks

 

I think I'll see if I can drop the pump speed and maybe open up the flows thro a few rads to keep the circulation rate the same

[John Carroll]

It depends on where the cold feed is teed into the system, if its just before the pump (suction) then the pump discharge head is the "height"+the pump head, in your case, 2M+3M=5M. If its after the pump on the discharge side then the pump suction head is the height-the pump head, 2M-3M=-1M, pump running with a negative head of 1M.

The rad in my converted attic runs with a very slight negative pressure, if you open the rad vent with the pump running then no water comes out but as soon as you stop the pump, water will start bleeding out, its run that way for ~ 30 years with no problems and heats up fully, as stated above, I require a pump head of 3.6M to fully heat all my rads.

[marshian]

9 hours ago, John Carroll said:

It depends on where the cold feed is teed into the system, if its just before the pump (suction) then the pump discharge head is the "height"+the pump head, in your case, 2M+3M=5M.

 

 

If its after the pump on the discharge side then the pump suction head is the height-the pump head, 2M-3M=-1M, pump running with a negative head of 1M.

 

Really good explanation and makes total sense

 

My Pump is above the cold feed (so 2.0 M plus 3.0 M Head = 5.0 M) so there shouldn't be much chance of air being drawn in - I'll continue my search for why I get occasional slugs of air in the circuits

 

I have turned the pump down from Speed 2 (3.0 M Head) to Speed 1 (1.8 M Head) to try and reduce the system noise and it all seems good on first pass (morning heating - selected rads)

[John Carroll]

1 hour ago, marshian said:

 

 

 

My Pump is above the cold feed (so 2.0 M plus 3.0 M Head = 5.0 M) so there shouldn't be much chance of air being drawn in - I'll continue my search for why I get occasional slugs of air in the circuits

 

I don't get that, you said, above that,  "I've got a pump head of 3.0 m but the actual height of the F&E tank water level to the pump is closer to 2.0 m" ?? can you provide a rough sketch also showing where the cold feed is teed in and position of the vent.

[marshian]

21 minutes ago, John Carroll said:

I don't get that, you said, above that,  "I've got a pump head of 3.0 m but the actual height of the F&E tank water level to the pump is closer to 2.0 m" ?? can you provide a rough sketch also showing where the cold feed is teed in and position of the vent.

Ok see below

 

[John Carroll]

That looks fine, flow from the boiler outlet (hot) with Vent then Cold feed then Pump?.

Sometimes the cold feed where its teed into the flow can get partially blocked and cause problems, can you watch the vent for pump over while someone statrs/stops the boiler/circ pump, also try holding that glass full of water with the vent end immersed in it to see if the level falls.

 

What type of filter is installed before the pump?.

[marshian]

14 minutes ago, John Carroll said:

That looks fine, flow from the boiler outlet (hot) with Vent then Cold feed then Pump?.

Sometimes the cold feed where its teed into the flow can get partially blocked and cause problems, can you watch the vent for pump over while someone statrs/stops the boiler/circ pump, also try holding that glass full of water with the vent end immersed in it to see if the level falls.

 

What type of filter is installed before the pump?.

Yeah I’ll try the glass of water trick

 

the filter is a magnaclean unit

[marshian]

Glass of water results (might have to repeat it with a larger glass)

 

pump on - level stable

 

pump off - load of bubbles followed by a level drop

 

pump on - all the water stolen from the glass

 

Just for info when on HW both cold feed and vent pipes below the pump get hot for a few feet up the wall so there is a level of water in them

 

When refilling the system from a drain down it fills fairly quickly so I think the cold feed is clear

 

I would have expected the vent pipe to have water in it to the same level as the F&E tank

 

 

 

[John Carroll]

1 hour ago, marshian said:

 

I would have expected the vent pipe to have water in it to the same level as the F&E tank

 

 

 

Yes and thats why the distsnce between the cold feed and the vent should be kept as little as possible, < 150mm so no pressure loss between the two. Its illogical allright but I have seen systems like yours that display the same problems "cured" by converting to my type, with the two combined as close to the F&E cistern as possible, needs a bit of plumbing though.

Edited Thursday at 08:48 by John Carroll

[marshian]

20 minutes ago, John Carroll said:

Yes and thats why the distsnce between the cold feed and the vent should be kept as little as possible, < 150mm so no pressure loss between the two. Its illogical allright but I have seen systems like yours that display the same problems "cured" by converting to my type, with the two combined as close to the F&E cistern as possible, needs a bit of plumbing though.

 

 

Ahh so one 22mm pipe feed from F & E to below the pump and then the vent tee's off close to the tank outlet - that would be do-able with existing pipework

 

Attached is picture of set up before recent changes to pump filter configuration and valving

[John Carroll]

Yes, something like that.

 

Re your existing set up as per your line drawing, can you give a estimate of the distances starting at the vent.... vent to cold feed, cold feed to filter and filter to pump inlet.

[marshian]

22 minutes ago, John Carroll said:

Yes, something like that.

 

Re your existing set up as per your line drawing, can you give a estimate of the distances starting at the vent.... vent to cold feed, cold feed to filter and filter to pump inlet.

 

They are effectively stacked (with the absolute min of pipe between all items)

 

Top

Zone Valves

Pipework

Gate Valve

Pump (130 mm flange to flange)

Gate Valve

Magnaclean Unit

22mm T (15mm inlet) Cold Feed

22mm equal T (22mm Vent Pipe)

22mm Flow from Boiler

Bottom

 

I can do exact measurements if required but it's all very tightly packaged

[John Carroll]

21 minutes ago, marshian said:

 

They are effectively stacked (with the absolute min of pipe between all items)

 

Top

Zone Valves

Pipework

Gate Valve

Pump (130 mm flange to flange)

Gate Valve

Magnaclean Unit

22mm T (15mm inlet) Cold Feed

22mm equal T (22mm Vent Pipe)

22mm Flow from Boiler

Bottom

 

I can do exact measurements if required but it's all very tightly packaged

So, the magnaclean, cold feed & vent are all very close to the pump inlet (suction end) and the arrow on the pump body is definitely pointing UP??

[marshian]

36 minutes ago, John Carroll said:

So, the magnaclean, cold feed & vent are all very close to the pump inlet (suction end) and the arrow on the pump body is definitely pointing UP??

Yes exactly that - 99.9% sure the pump body has an arrow pointing up but I'll check again

[John Carroll]

There should be a arrow on the filter body as well, check this is also pointing UP.

Is the filter clean?

Edited Thursday at 12:58 by John Carroll

[marshian]

1 hour ago, John Carroll said:

There should be a arrow on the filter body as well, check this is also pointing UP.

Is the filter clean?

 

I'm not aware of an arrow in the filter body but I'll check that too

 

Filter is checked every three months and rarely has much on it - below is as bad as it gets