15mm or 22mm pipe for CH? (or even 28mm)?

[Sam]

Hello,

I was browsing through some old posts and they said that 15mm pipe was suitable for carrying up to 6kW according to one post and 9kW according to another.

I'm a bit worried! It's very cold upstairs. We moved in last year and the first thing we did was to insulate the loft. That hasn't been enough, so I was going to change some radiators to bigger ones.

The thing is that the boiler has 22mm pipe come out of the boiler and upstairs but on the landing this is reduced to 15mm which tees, with one 15mm run going to the back rooms and one 15mm run going to the front rooms. I am sure there must be more than 6kW of radiator connected to that!

Is this why it's so cold? Will upgrading the rads. have no effect if it is that the pipe cannot carry that capacity?

I'm not sure if it would be easy to change the pipe work without ripping up the floor in each room. I would use plastic pipe to weave through the joists. At the moment they've taken a big chunk out the top of the joist to fit the two 15mm pipes through. That would be too small to fit two 22mm pipes in, so I guess I'd be best to drill two

22mm holes in the middle of the joists. How far away from the existing notch would I need to be to make sure I do not weaken the joist?

I also noticed that the gravity fed HW goes from the boiler in 28mm pipe but is reduced in the airing cupboard to 22mm.

In both these cases, doesn't it defeat the object of using big pipe if you then reduce it down later 22 to 15, or 28 to 22?

Thanks.

[fred]

In article , Sam writes

I'd say your first step is to balance the system, it is quite possible that this has not been done correctly and radiators closer to the boiler are stealing flow and therefore heat from the rest. Only if you cannot achieve a satisfactory balance do you need consider redesign and modification. The important thing is to go about it in a logical progressive fashion and not make random changes in the hope of improvement.

Here's a link for the uk.d-i-y group faq on radiator balancing:

useful; stuff here: here: the balancing section there is still a skeleton in draft

In brief, on the subject reducing pipe sizes, thing of the pipework as the a tree with a main trunk and smaller branches coming off the main, once the trunk is split there's no need to have a full size branch all the way to the top just to support a few leaves at the end.

[Bob Minchin]

Also check prior to balancing that each radiator is evenly warm from end to end and top to bottom with not cold patches. This could indicated sludge/partial blockages which need to be flushed out - possibly a summer job but at least knowing about might avoid expensive radiator upgrading.

Bob

[Sam]

Thanks. I could understand that but the reduction is size occurs before the branches. The 28mm goes from the boiler to the airing cupboard but before it gets to the cylinder changes to 22mm; there are no branches on that it runs to the cylinder only. The CH goes from the boiler in 22mm to the landing and splits into 15mm before that branches. My other worry is that once it's branches, posts here said

15mm supports only 6kW. Surely one big rad could use all that up?

[fred]

Do you mean a single 22 reduces to a single 15, I ask as you say splits into 15mm which suggests 22 splitting into 2 (or more) 15mm.

28mm dropping to 22mm for a short length is ok as (say) 8m of 28mm plus 2m of 22mm has less resistance to flow that 10m of 22mm.

The reduction to 15mm before the branches doesn't sound right but it may still be a serviceable design, try the balance and if that doesn't sort the problem then an upgrade to 22mm may be in order for a further part of the run.

Gravity fed H/W is not my forte but the heating comments should still be valid, I assume there is a separate pump and zone valve for the heating circuit.

[Roger Mills]

It's a bl**dy big rad that will produce 6kW of output. Tell us the size and construction (single, double, with/without fins) of each of your rads, and we will calculate the approx output of each.

You then need to calculate the heat losses of each room, and see whether the radiator capacity is adequate. You also need to calculate the *total* heat losses, plus an allowance for hot water to see whether the boiler is adequate.

If the design is ok, it is then simply(!) a matter of making sure that the system is balanced properly and is not sludged up.

In the rooms which are not warm enough, how hot are the rads getting? [Best measured with a infra-red thermometer but - failing that - for how long can you hold your hand on the rad? If you feel the inlet and outlet pipes, is the outlet a *lot* cooler than the inlet? Are the rads hot all over, or is there a cooler region in the centre towards the bottom?

How is the boiler stat set? If it's a gravity HW system, is there any control of HW temperature - other than by adjusting the boiler stat? If not, chances are that the boiler is set at only about 60 degC - whereas you need it at 80 to get more heat out of the radiators. [But then, the HW will get too hot!]

[Ed Sirett]

No one's mentioned, yet, what the temperature from the boiler is. There are such lot of basic simple stuff before we get into the subtle stuff.

1) Is the primary flow form the boiler touchable? 2) Are all the radiators hot? 3) Are all the radiators hot all over?

[Sam]

Thanks. Looks like I'm off to buy an IR thermometer then. Problem is, they cost as much as a small radiator ;(

[fred]

Get a feel for it by hand then, inlet should be hotter than you can bear to touch for any length of time, outlet should be less so, perhaps bearable for 15s (if you're butch). If any outlets are at the same or close in temp to the inlets then the rad needs throttling.

Obviously this is a very rough guide but it's a start and better than (potentially) nothing at all.

[Sam]

Hi. Thanks for the informative replies; I had so many, I didn't know which one to quote.

I have ordered an IR thermometer courtesy of a well-know auction site as this will be the quickest and easiest way to measure the temperatures and check the system is balanced. Like someone said, a little money spent on the thermometer now might save money on upgrades later.

I made a mistake by mixing my BTUs and kWs. The larger radiators at Screwfix have outputs of 6000ish BTUs, which is 2kW-ish; I was thinking they were 6kW radiators. My mistake, sorry. So I have more capacity in my pipe than I thought, which is good news.

The radiators in the house are old, single, without fins, and painted over. Heat outputs unknown.

It's getting a bit late, so I'll reply to Fred's questions in the morning. Good night.

[Roger]

The message from Sam contains these words:

Stelrad single panel?

23" high.

88" long 5921 Btus/hour.

82" 5498 76" 5075 69" 4652 63" 4229 57" 3808 50" 3384 44" 2961

[Roger Mills]

At what Delta-T value?

Those outputs look a bit optimistic to me - even at a delta-T of 60 degC (140 degF) my calcs come out a bit lower than that.

[Roger]

The message from "Roger Mills" contains these words:

Figures were taken from the Ravensbourne Guide to D.I.Y. Central Heating. It is not dated but I think it is perhaps circa 1975 judging by the prices and my contacts with the firm. It doesn't say how the figures are calculated but a later offering from Ravensbourne (probably circa

1988) does state "Manufacturers give emission figures for each radiator at a temperature difference between water and room of 60C or 108F".

The 88" radiator above gives 2.93 btus/sq. inch. The later catalogue give the closest Barlow radiator - 24*84 as 6105 which at 3.02 btus/sq. inch is higher but might not be significantly so as Stelrads dimensions include the seam top and bottom and the Barlow is a roll top. Ignore the seams and the average output is likewise 3.02.

[Fred]

Sorry I haven't replied to this thread for a while, I shall try to do so today or tomorrow. In the meantime can I ask one quick question:

Do I need 28mm pipe for central heating/hot water if the system is pumped? Is it that 28mm is only for gravity fed systems? If there is a pump in place, does 28mm offer any advantage over 22mm?

Thanks.

[Roger Mills]

28mm (or 1" Imperial in the 'old' days) is certainly needed for gravity circulation in order to minimise flow resistance - 'cos convection doesn't generate much pressure. 22mm is usually sufficient for pumped systems unless you have a very large or very leaky house with a correspondingly large boiler. There's no particular merit in using 28mm pipe unless you really need to. On the other hand, if you're converting a gravity system to pumped, and some conveniently located 28mm pipe is already installed, there's no point in *not* using it.

[Roger]

The message from Fred contains these words:

The larger the bore the more water can be pumped though and further but in a normal pumped system you may not need 28 mm even if you have one circuit serving the whole house.

My information is grossly out of date but the figures I have (taken from Ravensbourne Beginners Guide to Central Heating) are:

Length limits at maximum flow rates (3 feet/sec)

Tube size (mm) Max flow (gallons/hr Max length (feet)

15 105 80 22 240 120 28 390 175

And at those values the heat capacity and head loss are:

BTUs/hr inches water gauge/foot run

15 21,000 0.90 22 48,000 0.60 28 78,000 0.41