Balancing radiators - time taken for temperature change

I will be balancing radiators using a non-contact infrared thermometer as advised in

formatting link
how long does it take after either opening or closing lock shield valves for the temperature of the return water to change and stabilise? Would 10 minutes, for example, be about right, or does it take longer?

Reply to
Codswallop
Loading thread data ...

Try and see, but I suspect 2-3 minutes will be fine. Bear in mind that it has a small effect on all the other radiators too, so you will probably need to do a couple of rounds, at least to check.

I tended to do all mine in one go. Go round and check what the drop across each is. Work out how much each has to go up or down. Go round and adjust them all by an estimated amount tied to how far out they were. Wait a bit, and repeat, but second time you can see exactly what affect every change had first time, so you can base your estimate on that. Beware if any of your radiators form the bypass loop that you maintain enough flow through them when all the others shut off, and that may mean compromising on the temperature drop in that one case.

Reply to
Andrew Gabriel

An allied question: I recently had a go at balancing our radiators using the same info (thanks Phil!) and an IR thermometer from Maplin.

Given the time constraints I had I was mostly successful; I'm sure things are better balanced than before. However I wondered about the very last stages of the instructions, where you are instructed to adjust the control valves to reduce temperatures if necesary. I appreciate this may be necessary, but surely this will unbalance the system once again? You'll end up with mismatched impedances everywhere. Or is the assumption that (a) size of radiators in general are well-matched to the room size, & (b) Having to do this is a necessary evil?

Thoughtful ...

Jon N

Reply to
jkn

"jkn" wrote in news: snipped-for-privacy@e3g2000cwe.googlegroups.com:

You'll never get an accurate balance - be thankful with a lovely IR thermometer you can chase round after the points Andrew mentioned and get it nearly right!

The golden rule I have found out for myself, so it may be wrong, is never forget the "furthest" radiator you found, and makes sure it always stays wide open; I've been known to end up screwing them all down chasing my tail.

mike

Reply to
mike

In article , mike writes

I have my doubts, I'd never get a 20deg drop on a wide open rad unless it was fed by 20m of drinking straws.

Reply to
fred

fred wrote in news: snipped-for-privacy@y.z:

I think you would if your boiler is capable - you're looking for the same drop on all of them.

As you squeeze down the others, you will increase the drop on the coldest one.

If your boiler output can't drive all your radiators fully, you can still balance but not at the optimum temperatures

mike

Reply to
mike

In article , mike writes

Nup, still don't get it. The flow through the furthest radiator needs to be slow enough for the radiator to lose 20deg worth of heat by the time it passes through. If the pump's pressure hasn't been lost on the intervening pipework before it reaches that rad then the flow in that rad needs to throttled to ensure that drop, particularly if it is a small rad. If you squeeze down the others (flow?), the differential pressure rises and you get more flow through the distant rad resulting in a lower temp drop.

It just so happens that I look after a system where the distant rad is wide open when fully balanced but in my own that is not the case. Mine btw has separate 15mm feeds to each room so that may be a factor, there is no significant loading (self balancing) effect on any of my pipework.

I don't see where the boiler 'drive' comes into it.

Reply to
fred

"Codswallop" wrote: Thanks for all the replies. I don't expect to achieve perfect balance. The pump is upstairs and at the moment the bedroom radiators, which are nearest the pump, run hotter than the ones downstairs. I will be satisfied if I can get them all running at about the same temperature, or even the downstairs ones running a bit hotter than the upstairs ones.

Reply to
Codswallop

Forget about 20degrees (20F = 11C). Take another look at the Principles section of

formatting link
which finishes with this ...

"It doesn't matter whether you have that value or not when you balance as it can be adjusted by the boiler thermostat according to the heat you actually need. Similarly the oft quoted 11 deg C drop across the radiators is really nothing to do with the radiators! It is the temperature gradient the *boiler* is designed to produce when it is going flat out *and* the pump is producing the specified flow rate. So if you have less than 11C drop it just means that either you have too high a water circulation rate or/and the radiators are not large enough to extract the full rating from the boiler. This need not be a problem; it may be that the boiler is oversized for the house, and the converse of course follows."

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

True, but balance is really nothing to do with boiler output or particular temperatures. It simply adjusts resistances so that whatever heat happens to be available is distributed properly to all the radiators rather than finishing up in just a few of them. The temperature drop that the balanced radiators finish up with simply depends on (A) the boiler output, and (B) the total output rating of the radiators.

IF A = B you will have 11C for a conventional (i.e. older) system. For a system designed for a modern condensing boiler you could have nearer

20C.

However, it is very rare for A to be equal to B in practice, so it is pointless trying to achieve exactly 11C.

There is no 'optimum' temperature. However, if you designed your own system you would know what design temperature drop you worked on, and you would expect to see that figure after balancing.

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

If you get that, you HAVE achieved perfect balance :-)

The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

It's not wrong, it's a very good point!!

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

Your welcome - glad its helping. I'm taking note of questions in preparing for a clearer version.

That will be this bit in the FAQ...

"17. If necessary, adjust control valves to reduce room temperatures. This should only be necessary if a radiator is oversized and is thus over-heating that room, and does not apply if a thermostatic valve is fitted."

No, it won't cause a balance problem. Consider a system with no radiator valves at all. Radiators close to the boiler will hog the flow, short-circuiting distant ones which will not heat up. Balancing fixes this by increasing the resistance (or impedance in electrical terms) of the hog radiators so that more flow goes to the distant radiators.

Once balanced, closing a control (or hand, or TRV) valve on one radiator will increase the overall resistance of the circuit slightly, and since the pump continues to run at constant speed this will cause the system pressure to rise slightly. Result is that the remaining radiators will all get slightly more flow, but in equal proportions as they are still balanced with respect to each other.

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

Nooo, that's the wrong way round!! The flow through the furthest radiator needs to be FAST ENOUGH to get it hot. That is achieved by opening fully the furthest (most sluggish) one, and throttling back the others. As you throttling back the others, the sluggish one will perk up until it is as hot as all the others, with an equal (but arbitrary) temperature drop across them all.

There are 2 provisos in all this.

1) That the system pipework is adequate.

Sometimes you find installations with excessively long runs of pipe, or normal runs but undersized pipe bore. Be suspicious of this if you have a microbore system, or if no matter how hard you try one radiator just will not come up to full heat even with the valves fully open.

2) The pump is set to run too slowly or is undersized.

Start off with it at full speed, but don't forget after or during balancing to turn the it down to the lowest speed which gets the radiators piping hot. Excessive pump speed just causes noise in the pipes and valves, and premature wear out of TRVs. Once the radiators are up to full heat, you don't get any more heat however by running the pump faster.

No again.. You need to throttle the OTHER radiators to increase the flow in THIS one. See my previous re "Forget the 20deg"

Correct

Whoever set that up knew what they were doing.

You might be surprised how high the resistance of 15mm pipe is. There are tables telling you the max length it can be for particular loads. Unless it is a very compact system I would expect the main feed to be in

22mm, or even 28mm if its a big house or has the boiler at the far end.

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

Hi Phil

[some snipped]

Ah, OK. I think I see. I was thinking 'constant voltage' rather than 'constant current', which wouldn't help ;-/

Cheers Jon N

Reply to
jkn

I see where you're coming from but if it's an equal drop less than 20deg, how do I keep my boiler in condensing mode whilst maintaining maximum rad output :-?

It's an integrated modulating pump btw so I have no control over the setting.

Yes and no, the pipe design to the distant rad is barely adequate to feed the distant leg and the installer didn't bother to balance the system at all, one of those "it's got TRVs, it doesn't need balancing" types. It'll be no surprise to you that the warm up time was markedly reduced after I balanced it.

It's a long story but each room is a controlled zone with its own 15mm feed, when completed, each room can have different target temperatures at different times. Bedrooms warm morning and pre-bed, lounge evening only etc. It is also a compact shape with most runs having equal length.

Reply to
fred

That's a separate problem, and you won't solve it by mis-balancing. Best to search for Andy Hall's posts on the subject, but basically you want to maximize the drop across all radiators, not just one, whilst still having them produce the design output. A system designed for condensing delivers heat at a lower flow temperature, and has radiators with an oversized area so that they will still give the required output at the lower temperature.

Do you mean like the Grundfoss Alpha pump? I haven't noticed them referred to as modulating pumps, but on reflection its a fair description as it adjusts its flow rate according to demand. The basic principle of them is that as TRVs close off, the system pressure will start to rise (in same way the pressure at the end of a hosepipe rises when you put your thumb over it). To be honest I have not looked into these in detail so don't know how they are setup to the required design flow/pressure, or even if it is presetable.

Well, it should only be 'just' adequate ;-) but if it is not there are really only two options:

1) Increase the pump pressure. This can be achieved by increasing pump speed or replacing it with a more powerful one. Note that excessive pressure will have undesirable side effects like noisy flow, cavitation, premature failure of components. 2) Replace inadequate sections of pipework with larger bore pipes.

Or both!!

Sounds normal :(

Indeed, and presumably the output was back up to spec too.

That's fine. You might like to look at my post about balancing multiple zones in "Re: balancing radiators" earlier today. Bear in mind though that almost certainly there will be some times when all circuits will be on together, so the main feed from the boiler still needs to be adequate, and that probably means 22mm.

Radiator sizes for individually controlled rooms might need to be slightly larger than for whole house heating because heat can be lost to cooler adjacent rooms.

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

It's also worth pointing out that there isn't a sudden nirvana that happens when boilers are in "condensing mode". Recovery of latent heat does begin at the dew point, but in terms of efficiency overall, there is an increase of efficiency with reducing return temperature. At the dew point, there is an increase in the *rate* of change of efficiency not a sudden step change.

Therefore, the objective should be to minimise the return temperature by having adequately sized radiators for the required output. In turn, that allows the boiler to be run at a lower flow temperature.

In a way they are, but not in the same way that an integrated modulating pump is.

With an integrated pump the boiler controller controls the pump speed (might be in 3 or so discrete steps or analogue over a range) according to the boiler heat output.

The Alpha detects flow and head and adjusts its power accordingly according to resistance and a preset control.

Yes it is. In most cases, the Alpha can be left on its default setting, but there is an adjustment up and down from that.

Ultimately, if the pipe size to a radiator is inadequate for the heat output and hence flow required, neither balancing nor winding up the pump will help.

With balancing, if the pipes to the one radiator are inadequate, even turning off all the other radiators completely won't fix it.

With increasing pump output, ultimately there is increased turbulence in the pipes and apart from the noise, the flow doesn't increase to a useful level.

Certainly winding up the pump can help a bit within reason, but the ultimate fix if there is one radiator with obviously inadequate pipework is to replace the pipes or perhaps add a second radiator and derate the first.

Reply to
Andy Hall

You'll need to point out the bit where I suggested mis-balancing the system :-). My intent was to suggest that having the most distant radiator wide open is not the holy grail if the resulting balanced system has a drop that is less than a (condensing) boiler describes as optimum. If the drop on balance is only 10deg (I was talking Celsius btw) then I'd happily go round _all_ the rads throttling them down a tad to get a bit more drop whilst maintaining balance. That'll lower the return temp a bit and keep the boiler in its higher efficiency zone for a greater proportion of the time. It will mean however that the distant rad in not wide open.

I was meaning integrated into the boiler casing and modulated in that the speed was controlled by the boiler electronics based on the variable output of the (modulating) boiler. While they're basically a good idea it means that even an experienced user/designer has no control over pump flow.

Just not available on most modern systems I'm afraid, integrated pumps controlled by the boiler electronics.

Yes, but it'd be nice if the installer had designed it with a bit more headroom.

It is an established design, short 22mm feed and return manifolds with 8 off

2 port valves below the boiler.

Indeed, in practice I never have the unoccupied rooms set more than 4degs lower than heated ones but the rads were sized for the extra loss.

FAQ, it is much appreciated.

Reply to
fred

Ahh, nirvana, that's the word I was trying to recall to tell him to search for your description of it.

I knew that ;-) It's a one word name for the Alpha that's missing. Modulating is close but as you say not quite right.

Think you missed point 2) Andy ;-)

Phil The uk.d-i-y FAQ is at

formatting link
Google uk.d-i-y archive is at
formatting link
NOSPAM from address to email me

Reply to
Phil Addison

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.