Level of pipes in primary HW circuit

The fed to the coil in the cylinder doesn't really need to enter at the bottom connector you know, so you could lift the branch from the pump higher and place the zone valves on two different levels all together. That way you can keep the branch to the rad' circuit where it is, and only swap round the heating coil so it runs the other way. One TEE and a couple of elbows would do it.

- Take the flow from the boiler to the top of the airing cuboard and to the open vent of the F&E tank.

- Tee off at the top of the airing cuopboard and take the pipe down (the pump must be on this length half wat from the cylinder to the ceiling, as this reduces pump niose).

- Directly after the tee insert another tee and take this to the cold feed of the F&E tank.

- The two tees must be close together.

- The feed tee must be nearst to the pump.

- At the end of this downards length have a bend and insert the CH zone valve.

- Tee off just before the bend and insert the DHW zone valve, then to the top of the cylidner coil.

That is it. This is now self venting too and no need for manual air vents and puim pnoise vastly reduced. More space for maintenance too. Insert the pressure by-pass valve after the CH zone valve and it must be the last tee into the boilers return. Sorted.

Use adpters for the pump, not adpters and isolation valves. Put a full bore valve on:

- The flow just before the pump

- Just after the by-pass valve after the CH zone valve

- Just after the by-pass tee on the return.

Then with 3 valves you can isolate the cylinder, zone valbe, by -pass and pump.

I wouldn't bother. The "wet" part of a 2 port valve is usually pretty reliable. I wouldn't be convinced that the isolation valves would be any more so and there are 2 of those to go wrong!

Certainly, the replacement of the wet part of a 2 port valve is sufficiently infrequently required that a drain down is not an issue.

Christian.

Fair enough, but I still want full bore valves either side of the pump rather than the existing gate valve adapters - so I still need to raise the level of the tee above that of the existing 3-port valve. Even without isolation valves, I still can't get 2-port valves on the horizontal pipes either side of the tee.

So I still need to raise the pipework up higher and then come back down to the upper inlet on the cylinder (which is about half-way up).

Is this a problem?

Another thing: IMM is telling me that the output from the by-pass valve needs to last thing tee'd into the boiler return. Is this necessary? It would be much easier to tee it into the HW return before it disappears under the floorboards and joins up with the CH return somewhere or other. Is this a problem - bearing in mind that the by-pass will only operate when both zone valves are closed, so it can't make water flow backwards round the HW circuit.

It doesn't! The feed goes into the *top* connection - but this is only half way up the cylinder. The return comes out of the bottom connection which is only just above the floor.

You can get pump adapters with ball valves instead of gate valves. They should be considerably more reliable.

Go up to the T and then put the valves on the "back down" section. Place a bleed valve on the top of the loop. It is probably not necessary in use, though, if sealed pressurised operation is used, as it covers a multitude of installation sins. However you should still install the valve, in case it does prove necessary to use it.

Christian.

I have found that most pump adapters and integrated ball valves can't hack it. There are some good quality full bore versions around, but are difficult to get hold of.

Ah ha !!! With you now. So you really to make a loop that sits behind the hot water cylinder that will also act a by-pass loop for the whole system as well then. What orientation is the pump in? If it's standing up on end, with the inlet and outlet vertical, then you can still take a full loop of 22 mm copper straight up and make a big loop that sits behind the cylinder. You then take the indirect coil from a point along the riser pipe from the pump and connect it to the top most end of the coil the same as it is now.

With another TEE below the one for the water coil, you fit another zone valve for the central heating loop. One more TEE will take the return from the by-pass loop behind the cylinder back to the point where the return from the hot tank is now.

Would that work?

Just had another thought. Would it make it easier to install the DHW zone valve on the cylinder return connection? There's no reason it has to be on the flow side. Indeed, it will run 5-10C cooler on the return leg. The radiator valve can still stay on the flow side or also be moved to the return side provided no radiators have unusual return locations away from the main trunk.

Christian.

What the hell are you all on about? Just do what I suggested. Easy, low pump noises free flow, self venting, easy to isolate all parts in the airing cupboard, etc.

Well, your description isn't quite clear. In particular, you don't really explain where the fill connection goes. You mention a T for the pump before the T for the feed, but why would another T be needed? I'm hoping you mean just use the remaining outlet from the previous T for the feed. Perhaps a diagram would help?

In any case, there is no indication that the F&E tank pipework passes through this location, or that the system is gravity fed, in which case, other methods that use the space available may be required.

Finally, the system proposed requires the boiler to have secondary overheat protection. If this is available, you may as well have combined vent/feed as well, to eliminate pumping over problems, or better still, convert to sealed pressurised operation.

Note that if the boiler does not have secondary overheat protection (normally indicated by suitability for sealed/pressurised operation), then the proposed control layout must be changed. This is because there needs to be a direct unvalved connection between the header feed and the boiler return that does not cross over or combine with the venting arrangements. This is because the boiler needs a supply of quenching water to prevent it boiling dry. Rising steam, or an automatic bypass valve would prevent such supply.

Christian.

I decided to post a couple of pictures to make clear what I'm on about!

The first one

The system is currently vented, although I may convert it to sealed at some time. The boiler (Baxi Solo MK I) *is* suitable for sealed operation. The F&E tank is in the attic directly above the airing cupboard. There is a single 22mm pipe tee'd into the flow pipe just under the floor and running across to the LHS of the airing cupboard, where it splits into 2 pipes which go up the wall to the attic - not visible in the picture - 22mm vent and

The second picture

• get rid of the 3-port valve and the 2 horizontal pipes either side of it
• extend the pump outlet pipe up higher (probably raising the pump at the same time to get a full bore valve under it
• run 2 horizontal pipes from the top of the extended flow pipe - one going down through a 2-port valve to the top coil connection, and the other going down through another 2-port valve to connect into the CH flow pipe
• fit a vent at the highest point (the top right red bit)
• fit an automatic by-pass valve between the flow pipe above the pump and the HW return - replacing the existing gate valve with a tee (I can fit a gate valve on the inlet instead, if needed)

Moving things up like this gives me a lot more room to manoeuvre - including space for a few full-bore isolation valves to enable component changes without draining

Are there any glaring gotchas in my proposal?

Thanks for your interest.

Again:

The boiler is downstairs, the F&E tank in the loft. The flow from the boiler will go directly to the overflow over the F&E tank.

- In the airing cupboard tee off as high as possible near the ceiling on this 22mm pipe.

- Bend down to the cylinder beneath.

- Tee off, to top of cylinder coil

- This pipe then enters the 2-way valve.

- a bend from the tee at the cylinder and insert the ch 2-way valve

- half way down the pipe from the top tee to the 3-way valve tee insert the pump.

You will have a straight run either side of the pump reducing water noises. Aso, it will make the system self venting. Any air in the DHW coil will rise up through the 2-way valve past the pump and to the vent. Same with the CH side.

To stop the pump sucking in air, take the 15mm cold feed from the F&E tank and tee it into the pipe from the open vent pipe tee to the pump. Keep it just before the tee, in other words the two tees together. The pump pulls down on the cold feed from the F&E tank and open vent, with virtually the same suction on both

- Self venting.

- Low noise

- no air sucked in.

Draw it out. It is one simple way of preventing air accumulating and reducing pump noises. The pump will also be above the cylinder and easy to get at.

Use fixed font.

============ Vent || ||¦ ¦ ||¦ F&E tank ¦ || ----------- || ¦¦ || ¦¦T (cold feed from F&E) T||===== Horiz. pipe near linen || || cupboard ceiling T = Tee || C|| || || || C|| Pump in centre of || (P ) long straight firmly || C|| clipped (C) to wall || || || C|| || ||T || ZV====ZV= -> To DHW coil flow |||| (ZV = Z valve) |||| || Flow to CH rads || From boiler flow (Combined fill, vent & flow)

Personally, I would use the automatic bypass valve to "short out" the hot water zone valve. I would also remove the balancing gate valve. Hopefully, this will give hot water priority and the pump overrun will dump any heat into the cylinder. This is NOT an appropriate solution on an unvented cylinder. It should only be considered for gravity cylinders and heat banks.

Christian.

Tha't an interesting idea. Dumping a bit of heat into the HW for a few seconds (during pump over-run) certainly wouldn't do any harm.

Presumably you are referring to unvented *cylinders*? I assume that if I convert the *primary* circuit to unvented, that wouldn't be a problem?

It's probably a good idea to get rid of the balancing gate valve - in fact, I'll open it right up now to see what happens. Since I've fitted a programmable stat on the CH, I generally heat the HW when the CH isn't on, anyway - so there's no need for any balancing.

Can you see any other problems in what I want to do - with reference to the pictures mentioned in my earlier post? [

There is one disadvantage I can think of, though. You could possibly overheat the water if all the TRVs are set low and the ABV starts to open. I don't know if this would cause much of a problem in a cylinder. It would cause none in a heat bank, as the water is supposed to be kept hot and it is water that would have passed through anyway.

Yes, an unvented cylinder. The primary can be sealed pressurised.

Christian.

Having an auto by-pass dumping heat into a cylinder is a no, no, unless. the cylinder has a boiler high temp cut off and a hot water draw-off blending valve. Without these, water in the cylidner may get to scalding level and the cylinder will may get to boiler flow temperature (not a problem if a blending valve is used).

Yes, that's why I would suggest the heat bank might be more suitable for it. These are typically fitted with blending valves or other systems of temperature control.

Christian.

Back to the main subject, if I can't dump the bypass into the cylinder, what's wrong with dumping it into the HW return before it merges with the CH return, as per