Does it matter which way water circulates in an indirect hot water cylinder coil

[Tim Snell]

Right... long time no posting,

I hope that title gave the right impression.

Over the past week I have had to replace an old indirect HW cylinder due to a number of leaks.

I have replaced the old (900x450 - 120L) with a larger one (1200x450 -

162L). It was pretty much a straight swap apart from changing the shower pump connection from an essex flange to a surrey flange. Certainly I have not changed the layout of pipes to the coil from the boiler.

The system is a 1992 vintage Johnson Starley warm air unit 'J50' or something like that with a built in Eljan water heater. The water heater is used exclusively for indirect heating of the water cylinder. The system is 'gravity fed' with the cylinder directly above the bolier displaced by about 2m horizontally. unlagged 22mm copper throughout

There were some fun and games along the way, most notably a stubborn airlock in the coil which I cleared using mains pressure via a hose into the vent pipe in the loft. The system is now 'working' in that I have hot water. However I had never really examined the old system prior to the last week but since giving my undivided attention in the evenings I have noticed something, I think, is wrong.

Flow comes from the bolier into the coil/heat exchanger at the bottom of the cylinder and exits at the top before returning to the boiler.

I think this seems wrong (detrimental to gravity feed, heat exchanger ideally should be hottest at top, flow pipework from boiler needs to be very hot), the diagram here suggests it is wrong.

have a few questions

Q1 Is there any way the flow in a gravity fed system can get reversed? Sounds stupid I know but how is the flow normally encouraged in one direction - any valves or just position and layout of pipes and heat exchanger

Q2 Bearing in mind the system seems to work now (and I have *not* changed the setup) how bad is it to have it plumbed like this? Inefficient, slower to heat tank, dangerous.

Q3 Is this 'mistake' common, are there any legitimate reasons for doing it this way and should I change it or am I worrying about nothing

Thanks Tim

[Grimly Curmudgeon]

We were somewhere around Barstow, on the edge of the desert, when the drugs began to take hold. I remember Tim Snell saying something like:

Inefficient, certainly; slower to heat the tank, yes; dangerous, probably not. I've seen a couple of cack-handed installations that worked, but depended on a great roaring fire to get flow going and were always on the verge of kettling in the back boiler. One of them was as you described and the other, while plumbed in correctly as regards flow, was damn near defiant of the law of gravity.

If it were me, I'd re-plumb the coil connections.

[TheScullster]

"Grimly Curmudgeon" wrote

If you do, add a bleed at the highest point to get rid of air in the coil (if your system doesn't have one already)

Phil

[dom]

You might find the solid fuel Rayburn installation instructions some use:

[RobertL]

Why not lag the upgoing pipe, especially right near the boiler. That should help get the flow going more quickly (and in he right direction).

Robert

[Tim Snell]

When you say 'as you described' what do you mean? I can not categorically say that it is plumbed incorrectly as I do not really know which 22mm pipe coming out of the boiler is the flow and which the return. One has a draincock attached and the other has a probe thermocouple phial in it.

So perhaps the water is now stubbornly going the wrong way in which case I want to address this rather than the 'quick' (anything other given the access restrictions) fix of replumbing the coil

I am inclined to agree - which is why I posted in the first place really I just wanted to gauge opinion.

[Tim Snell]

I have made sure that all pipe is angled up to the vent pipe so hopefully this will achieve the same result

[Tim Snell]

Once issues are resolved I intend to lag the bits I can see. As posted above to be effective it does require very high temp in the boiler and indirect pipework which is not pleasant in the summer.

[Tim]

Given that the boiler is the heat source and that hot water is less dense than cold, in a gravity system the top pipe on the boiler has pretty much got to be the "flow" and the bottom the "return".

The reason for the flow being connected to the top of the cylinder coil is so that there is always a temperature gradient between the coil and the surrounding water, maximising heat transfer. The water from the bottom of the coil will be as cold as it can possibly get in such a system which will improve its return to the boiler and hence gravity flow through the system.

You could add a pump whilst you're there. Although gravity systems work, a fully pumped system works an awful lot better. Depending on your hot water demands, you could notice a vast improvement in your cylinder's "recovery" time.

Tim D

[Heliotrope Smith]

Tim Snell wrote in news: [e-mail]:

The pipe with the thermostat pocket is the flow and the lower pipe with the drain off is the return. You should be able to trace the flow pipe to the top connection on your cylinder. If this is not teed off to vent into a feed/expansion tank then you are most likely to have a direct cylinder.

Are you sure you have an indirect cylinder or do you mean "direct" which doese not have a coil which is the usual set up for your hot water circulator type?

Do you have a large and a small tank in the loft or just a large tank? (It is not unknown to have indirect cylinders on these but is not very common.)

--- news://freenews.netfront.net/ - complaints: [e-mail] ---

[Tim Snell]

[e-mail]:

I am absolutely sure it is indirect! I replaced the cylinder and plumbed the coil.

Many of the houses on the 'estate' built circa 1970 with warm air systems have later been converted with a similar heater to mine but with a primatic cylinder. Mine, however, is definitely indirect.

[Tim Snell]

I would like to add a pump but I am not sure how I would 'control' this. In previous CH systems I have owned the boiler took care of the when to start stop the pump based on feedback from thermocouples etc. Very crudely I could probably add a pump so that when gas is burned the pump circulates but normally you would want it to run on for a bit.