And another question for the assembled experts ...

I'm building a 3 zone heating system and plan to fit gate valves to
allow a rough balance between the zones. I'd intended to fit them after
the zone valves but someone has just said that they should be on the
return side - does it make any difference whether they're on flow or
return?
--
Dave S
(The return email address is a dummy)

I'm building a 3 zone heating system and plan to fit gate valves to
allow a rough balance between the zones. I'd intended to fit them after
the zone valves but someone has just said that they should be on the
return side - does it make any difference whether they're on flow or
return?

I would balance the radiators individually with all zones open. I would not
balance the zones as a whole.

Christian.

Dave wrote:
they should be on the
return side - does it make any difference whether they're on flow or
return?

Normal (commercial) practice is to fit a double regulating valve on the
return, with an isolating valve on the flow. The only reason for the
regulating valves on the return is that they're subject to lower
temperatures, so it doesn't make any difference. The DRVs usually
incorporate a flow measuring device.

Gate valves aren't much good for regulating, most of the flow
regulation occurs in the last turn & they can generate noise. You'd do
better to get some proper DRVs. They show up on E-bay occasionally and
rarely attract any bids. A manifold with built in valves & flow
indicators would probably be best of all, but the flow indicators
become illegible after a year or two.

Aidan wrote:
Dave wrote:

they should be on the
return side - does it make any difference whether they're on flow or
return?


Normal (commercial) practice is to fit a double regulating valve on the
return, with an isolating valve on the flow. The only reason for the
regulating valves on the return is that they're subject to lower
temperatures, so it doesn't make any difference. The DRVs usually
incorporate a flow measuring device.

Gate valves aren't much good for regulating, most of the flow
regulation occurs in the last turn & they can generate noise. You'd do
better to get some proper DRVs. They show up on E-bay occasionally and
rarely attract any bids. A manifold with built in valves & flow
indicators would probably be best of all, but the flow indicators
become illegible after a year or two.


Thanks, I hadn't heard of these.

--
Dave S
(The return email address is a dummy)

Dave wrote:

Thanks, I hadn't heard of these.

Double Regulating Valves look like this;

http://cgi.ebay.co.uk/ws/eBayISAPI.d...sPageName=WD2V

This ebay sale is finished, only the pic is relevant. The two small
connections are to attach a differential pressure meter, used to
measure flow rates. Not required for domestic balancing. They're called
'double' because the setting can be locked. The DR Valve can then be
shut to isolate flow, but can only be opened as far as the locked
setting.

On Tue, 1 Mar 2005 13:21:25 -0000, in uk.d-i-y "Christian McArdle"
wrote:

I'm building a 3 zone heating system and plan to fit gate valves to
allow a rough balance between the zones. I'd intended to fit them after
the zone valves but someone has just said that they should be on the
return side - does it make any difference whether they're on flow or
return?

I would balance the radiators individually with all zones open. I would not
balance the zones as a whole.

Agreed, with the assumption that all zones are to be operated together,
at least some of the time.

There is no benefit in attempting to balance the zones the way you
suggest, Dave, as the objective is to get the correct flow through each
radiator and that means restricting the individual radiator flows. Once
that is done you have finished!! There is no point in restricting any
flow any further, indeed that would simply starve that zone.

As to whether LSV's go in flow or return, it does not matter either. Any
restricting device will work equally well in the flow as the return -
there is really no difference, its just a continuous circuit. The only
'difference' between flow and return is that the water is a bit cooler
by the time it reaches the return to the boiler.

Phil
The uk.d-i-y FAQ is at http://www.diyfaq.org.uk/
The Google uk.d-i-y archive is at http://tinyurl.com/65kwq
Remove NOSPAM from address to email me

On 1 Mar 2005 12:02:26 -0800, in uk.d-i-y "Aidan"
wrote:


Dave wrote:

Thanks, I hadn't heard of these.

By the way, Dave, you have read the Balancing FAQ haven't you?

Double Regulating Valves look like this;

http://cgi.ebay.co.uk/ws/eBayISAPI.d...sPageName=WD2V

This ebay sale is finished, only the pic is relevant. The two small
connections are to attach a differential pressure meter, used to
measure flow rates. Not required for domestic balancing. They're called
'double' because the setting can be locked. The DR Valve can then be
shut to isolate flow, but can only be opened as far as the locked
setting.

That's interesting. I've often thought there should be an absolute way
of setting the correct flow rates. Those DRV's, though ugly, are
presumably much quicker to set up than the iterative temperature drop
method.

Phil
The uk.d-i-y FAQ is at http://www.diyfaq.org.uk/
The Google uk.d-i-y archive is at http://tinyurl.com/65kwq
Remove NOSPAM from address to email me

"Dave" wrote in message
...
And another question for the assembled experts ...

I'm building a 3 zone heating system and plan to fit gate valves to
allow a rough balance between the zones. I'd intended to fit them after
the zone valves but someone has just said that they should be on the
return side - does it make any difference whether they're on flow or
return?

This could be noiser than using the valves on each radiator. I do install a
valve on the return from each circuit (with the zone valve on the send) but
only so I can drain each one individually if needed.

In article ,
Phil Addison writes:

As to whether LSV's go in flow or return, it does not matter either. Any
restricting device will work equally well in the flow as the return -
there is really no difference, its just a continuous circuit. The only
'difference' between flow and return is that the water is a bit cooler
by the time it reaches the return to the boiler.

LSV (or rather, flow restriction) in the return is quieter for the
week or so after the system has been filled when there are still air
bubbles working their way out. They can be quite noisey going through
the flow restrictor into a radiator, but of course they don't leave
the radiator in the return pipework, having separated out.

BTW, I noticed my Dryton TRV's all also have the flow restrictor
balancing adjustment in them. You need a special tool which I don't
have and the instructions are rather vague. It says something like
you can only turn the restrictor one way to adjust it, and it's not
very clear if having closed it down, you can actually decide to open
it any afterwards. Maybe if you carry on turning it flips back to
fully open if you go all the way round? Has anyone used this feature
of the Dryton TRVs?

--
Andrew Gabriel

Phil Addison wrote:
On 1 Mar 2005 12:02:26 -0800, in uk.d-i-y "Aidan"
wrote:
.... snipped

By the way, Dave, you have read the Balancing FAQ haven't you?

I have, but I'll read it again.

I've got 3 CH zones (z1=5+2 rads on different runs, z2=7 rads, z3=4
rads), one zone for DHW, the zones have very different pipe runs and the
DHW cylinder is next to the boiler. It should be fairly easy to balance
rads within a zone but, if I balanced just with the lockshields, I
thought that the different zone impedances would mess-up an overall
balance and the zones with the lowest impedance would dominate until
their TRVs shut down. By equalizing the zones I should be able to get a
better balance.

Where is my thinking screwed-up?

.... snipped

--
Dave S
(The return email address is a dummy)

On Wed, 2 Mar 2005 09:44:14 +0000 (UTC), in uk.d-i-y
(Andrew Gabriel) wrote:

In article ,
Phil Addison writes:

As to whether LSV's go in flow or return, it does not matter either. Any
restricting device will work equally well in the flow as the return -
there is really no difference, its just a continuous circuit. The only
'difference' between flow and return is that the water is a bit cooler
by the time it reaches the return to the boiler.

LSV (or rather, flow restriction) in the return is quieter for the
week or so after the system has been filled when there are still air
bubbles working their way out. They can be quite noisey going through
the flow restrictor into a radiator, but of course they don't leave
the radiator in the return pipework, having separated out.

That's a point, but I don't think a couple of weeks of hissing valves
out vote the advantage of having control (or TRV) valves on the most
convenient end for the householder. There are often situation where a
radiator is mostly behind some piece of furniture, and one end is much
easier/convenient for the end user to reach than the other.

Phil
The uk.d-i-y FAQ is at http://www.diyfaq.org.uk/
The Google uk.d-i-y archive is at http://tinyurl.com/65kwq
Remove NOSPAM from address to email me

On Wed, 02 Mar 2005 11:14:49 +0000, in uk.d-i-y Dave
wrote:

Phil Addison wrote:
On 1 Mar 2005 12:02:26 -0800, in uk.d-i-y "Aidan"
wrote:
... snipped

By the way, Dave, you have read the Balancing FAQ haven't you?

I have, but I'll read it again.

I've got 3 CH zones (z1=5+2 rads on different runs, z2=7 rads, z3=4
rads), one zone for DHW, the zones have very different pipe runs and the
DHW cylinder is next to the boiler. It should be fairly easy to balance
rads within a zone but, if I balanced just with the lockshields, I
thought that the different zone impedances would mess-up an overall
balance and the zones with the lowest impedance would dominate until
their TRVs shut down. By equalizing the zones I should be able to get a
better balance.

Where is my thinking screwed-up?

Dunno! But try envisaging the same radiators without the desire to zone
them. In that case balancing works just fine as normal, and all that
adding ZVs does is to shut off some rads when they are not needed. There
is nothing magic about a zone - it is just a group of rads that you want
to be able turn off with one valve (instead of running round shutting
off each rad's control valve).

Phil
The uk.d-i-y FAQ is at http://www.diyfaq.org.uk/
The Google uk.d-i-y archive is at http://tinyurl.com/65kwq
Remove NOSPAM from address to email me

There is a morsel of logic in there somewhere.

Commercial system balancing is done by measuring the flow rates and
proportionally balancing the system.

If you call the furthest heater/rad No.1 & the second furthest No. 2,
then you'd measure flow rates on No.1 and No.2. Then you'd reduce No.2
flow rate until 1 & 2 are the same percentage of the design. At the
start No.1 might be 45% & No. 2 might be 70% of the design flow rate.
You'd reduce No.2 flow rate & this would cause No.1 to increase. When
you're done they might both be 55%.

Then you do the same with No. 3 & No. 2. This time, as you reduce No.3
flow, Nos.1 & 2 flows will increase in the same proportion. Then No.4 &
No. 3, etc., etc., ad nauseam.

You'd then balance all the zones using the zone valves (DRVs) and
finally set the system flow rate using a DRV on the pump outlet.

You could apply this logic using return temperatures.

Any restriction will generate some noise. The noise is less likely to
be intrusive if some of the restriction is done at zone valves, rather
than all at the rad valves.

On 2 Mar 2005 10:46:38 -0800, in uk.d-i-y "Aidan"
wrote:

There is a morsel of logic in there somewhere.

You mean in Dave's post?

Commercial system balancing is done by measuring the flow rates and
proportionally balancing the system.

If you call the furthest heater/rad No.1 & the second furthest No. 2,

"Furthest" being in the hydrodynamic sense, presumably, since a small,
close radiator, fed by a narrow bore may well be 'further' away in terms
of resistance than a beefy rad which is physically further away, fed by
a large bore pipe?

then you'd measure flow rates on No.1 and No.2. Then you'd reduce No.2
flow rate until 1 & 2 are the same percentage of the design. At the
start No.1 might be 45% & No. 2 might be 70% of the design flow rate.
You'd reduce No.2 flow rate & this would cause No.1 to increase. When
you're done they might both be 55%.

Surely this example is true only if No.1 and No.2 are the same loading?

Then you do the same with No. 3 & No. 2. This time, as you reduce No.3
flow, Nos.1 & 2 flows will increase in the same proportion. Then No.4 &
No. 3, etc., etc., ad nauseam.

Basically you are trying to adjust each radiator circuit's resistance
until they all take their design percentage of the total system load
(for that zone), yes?. This is the same aim as the temperature drop
method, and the iterative procedure seems to be nearly as 'ad nauseous',
as you note. I see a couple of advantages though; 1) you don't have to
wait for temperatures to stabilise, and indeed you don't even need to
fire up the boiler. 2) Reading off a pressure gauge should be quicker
and more repeatable than trying to get an accurate temperature reading.

You'd then balance all the zones using the zone valves (DRVs) and
finally set the system flow rate using a DRV on the pump outlet.

You could apply this logic using return temperatures.

Well, yes - it is the exact same concept. We DIYers only use temperature
drop because we don't have the luxury of flow measurement. I have
thought long about a simple non-intrusive method of measuring flow rates
inside central heating pipes, but with singular lack of success.

Any restriction will generate some noise. The noise is less likely to
be intrusive if some of the restriction is done at zone valves, rather
than all at the rad valves.

True, but I reckon most domestic c/h need all the flow they can get,
what with tubing and pumps sized to minimise cost. Should you find the
flow is excessive, the domestic answer is to turn the pump down.

Thanks for the interesting insight into large scale techniques.

Phil
The uk.d-i-y FAQ is at http://www.diyfaq.org.uk/
The Google uk.d-i-y archive is at http://tinyurl.com/65kwq
Remove NOSPAM from address to email me