Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Central Heating header tank location/water head
380 views
Skip to first unread message
Tim Watts
Dec 29, 2012, 8:09:15 PM12/29/12
to
Just surveying positions for the header tank...
(Please don't say sealed system - I am committed to a vented system)
Now, most of the CH is downstairs and I have a nice position for the header
tank such that the base (and thus water head) is 150cm *above* the ground
floor ceiling, and about 130cm above the upstairs floor.
The pumps[1] will be 30-40 cm below the ceiling
[1] Boiler circuit, rads, HW plate exchanger all have sperate pumps
into/from a thermal store. The header tank feeds the thermal store -
basically just a 150 litre copper cylinder with extra tappings.
Am at any risk of lack of head possibly leading to pump over problems?
There will probbaly be 2 rads upstairs, short profile, so tops > 50cm under
the bask of the header tank.
Second question - is it usual these days to have a seperate feed to CH from
the header tank and a seperate vent pipe hanging over the top - or a single
feed in 22mm - with downward slopes of course, all the way?
I *could* get the tank right up to the ceiling upstairs, but it's a bit
awkward for other reasons - this is a hipped single story bungalow with
dormers upstairs. Hanging the tank inside the top of the dormer is tricky.
I am hoping to fit it in a cubby hole at the top of the stairs unde rthe
hipped roof which puts it more or less over the thermal store.
This will meet the boiler's typical demands for 2m head minimum given the
chose position for the boiler.
Cheers!
Tim
--
Tim Watts Personal Blog: http://www.dionic.net/tim/
"She got her looks from her father. He's a plastic surgeon."
(Please don't say sealed system - I am committed to a vented system)
Now, most of the CH is downstairs and I have a nice position for the header
tank such that the base (and thus water head) is 150cm *above* the ground
floor ceiling, and about 130cm above the upstairs floor.
The pumps[1] will be 30-40 cm below the ceiling
[1] Boiler circuit, rads, HW plate exchanger all have sperate pumps
into/from a thermal store. The header tank feeds the thermal store -
basically just a 150 litre copper cylinder with extra tappings.
Am at any risk of lack of head possibly leading to pump over problems?
There will probbaly be 2 rads upstairs, short profile, so tops > 50cm under
the bask of the header tank.
Second question - is it usual these days to have a seperate feed to CH from
the header tank and a seperate vent pipe hanging over the top - or a single
feed in 22mm - with downward slopes of course, all the way?
I *could* get the tank right up to the ceiling upstairs, but it's a bit
awkward for other reasons - this is a hipped single story bungalow with
dormers upstairs. Hanging the tank inside the top of the dormer is tricky.
I am hoping to fit it in a cubby hole at the top of the stairs unde rthe
hipped roof which puts it more or less over the thermal store.
This will meet the boiler's typical demands for 2m head minimum given the
chose position for the boiler.
Cheers!
Tim
--
Tim Watts Personal Blog: http://www.dionic.net/tim/
"She got her looks from her father. He's a plastic surgeon."
Dave Plowman (News)
Dec 30, 2012, 4:58:13 AM12/30/12
to
In article <rqg4r9-...@squidward.local.dionic.net>,
sealed system for warranty purposes.
--
*What do you call a dinosaur with an extensive vocabulary? A thesaurus.*
Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.
Tim Watts <tw+u...@dionic.net> wrote:
> Just surveying positions for the header tank...
> (Please don't say sealed system - I am committed to a vented system)
Have you checked the boiler manufacturer is ok with this - some require a
> Just surveying positions for the header tank...
> (Please don't say sealed system - I am committed to a vented system)
sealed system for warranty purposes.
--
*What do you call a dinosaur with an extensive vocabulary? A thesaurus.*
Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.
Tim Watts
Dec 30, 2012, 5:18:58 AM12/30/12
to
Dave Plowman (News) wrote:
> In article <rqg4r9-...@squidward.local.dionic.net>,
> Tim Watts <tw+u...@dionic.net> wrote:
>> Just surveying positions for the header tank...
>
>> (Please don't say sealed system - I am committed to a vented system)
>
> Have you checked the boiler manufacturer is ok with this - some require a
> sealed system for warranty purposes.
>
Hi Dave,
> In article <rqg4r9-...@squidward.local.dionic.net>,
> Tim Watts <tw+u...@dionic.net> wrote:
>> Just surveying positions for the header tank...
>
>> (Please don't say sealed system - I am committed to a vented system)
>
> Have you checked the boiler manufacturer is ok with this - some require a
> sealed system for warranty purposes.
>
I'll choose a boiler that is compatable. Viessmann are fine - read one of
their manuals. They do however stipulate a 2m head at the boiler which I
took into account.
Pretty sure Worcester-Bosch are OK too.
"It would be better to live under robber barons than under omnipotent
moral busybodies."
Dave Plowman (News)
Dec 30, 2012, 6:11:05 AM12/30/12
to
In article <i1h5r9-...@squidward.local.dionic.net>,
no open header tank. But it's not a combi but a 'system' boiler so comes
with internal pressure vessel anyway.
Can I ask why you don't want a pressurized system? I'm a bit of a dinosaur
about such things but happily changed my system when fitting the
Viessmann, and have no regrets. Indeed it is much easier to bleed - the
old system was a bit of a pain at initial fill with airlocks.
I'll dig out the installation instructions.
--
*INDECISION is the key to FLEXIBILITY *
Tim Watts <tw+u...@dionic.net> wrote:
> > Have you checked the boiler manufacturer is ok with this - some
> > require a sealed system for warranty purposes.
> >
> Hi Dave,
> I'll choose a boiler that is compatable. Viessmann are fine - read one
> of their manuals. They do however stipulate a 2m head at the boiler
> which I took into account.
I've got a Viessmann, and that's why I asked, as I'm pretty sure it said
> > Have you checked the boiler manufacturer is ok with this - some
> > require a sealed system for warranty purposes.
> >
> Hi Dave,
> I'll choose a boiler that is compatable. Viessmann are fine - read one
> of their manuals. They do however stipulate a 2m head at the boiler
> which I took into account.
no open header tank. But it's not a combi but a 'system' boiler so comes
with internal pressure vessel anyway.
Can I ask why you don't want a pressurized system? I'm a bit of a dinosaur
about such things but happily changed my system when fitting the
Viessmann, and have no regrets. Indeed it is much easier to bleed - the
old system was a bit of a pain at initial fill with airlocks.
I'll dig out the installation instructions.
--
*INDECISION is the key to FLEXIBILITY *
Dave Liquorice
Dec 30, 2012, 6:37:42 AM12/30/12
to
On Sun, 30 Dec 2012 01:09:15 +0000, Tim Watts wrote:
> Am at any risk of lack of head possibly leading to pump over problems?
Think about the static water levels in the tank and pipes. The water
> Am at any risk of lack of head possibly leading to pump over problems?
level in the open vent pipe will be the same as that in the tank. You
only need enough "pressure" to lift the water up to the base of the loop
over the top of the tank to get pump over. The head doesn't really come
into it.
> Second question - is it usual these days to have a seperate feed to CH
> from the header tank and a seperate vent pipe hanging over the top - or
> a single feed in 22mm - with downward slopes of course, all the way?
the boiler, CH and DHW heat exchanger. I'd fit the feed and expansion as
per normal to the tank via their own tappings if possible. ie feed to the
bottom and expansion always rising from the very top of the cyclinder. A
single pipe would have to be connected at the very top to vent any
air/gases evolved in the store. When the store cools it'll draw
relatively cold water from the header tank into the top of the store. It
won't stay there, being denser, it will sink setting up circulation that
will tend to destroy any stratification in the store.
Note that the feed and vent pipes are now not in any pumped circuit so
pump over (or draw down) might not be such a big issue but I'd still make
the vent loop over the top of the haeder a "sensible height".
--
Cheers
Dave.
Tim Watts
Dec 30, 2012, 7:45:57 AM12/30/12
to
Dave Plowman (News) wrote:
> In article <i1h5r9-...@squidward.local.dionic.net>,
> Tim Watts <tw+u...@dionic.net> wrote:
>> > Have you checked the boiler manufacturer is ok with this - some
>> > require a sealed system for warranty purposes.
>> >
>
>> Hi Dave,
>
>> I'll choose a boiler that is compatable. Viessmann are fine - read one
>> of their manuals. They do however stipulate a 2m head at the boiler
>> which I took into account.
>
> I've got a Viessmann, and that's why I asked, as I'm pretty sure it said
> no open header tank. But it's not a combi but a 'system' boiler so comes
> with internal pressure vessel anyway.
There's another range that is fine for open vented.
> In article <i1h5r9-...@squidward.local.dionic.net>,
> Tim Watts <tw+u...@dionic.net> wrote:
>> > Have you checked the boiler manufacturer is ok with this - some
>> > require a sealed system for warranty purposes.
>> >
>
>> Hi Dave,
>
>> I'll choose a boiler that is compatable. Viessmann are fine - read one
>> of their manuals. They do however stipulate a 2m head at the boiler
>> which I took into account.
>
> I've got a Viessmann, and that's why I asked, as I'm pretty sure it said
> no open header tank. But it's not a combi but a 'system' boiler so comes
> with internal pressure vessel anyway.
> Can I ask why you don't want a pressurized system? I'm a bit of a dinosaur
> about such things but happily changed my system when fitting the
> Viessmann, and have no regrets. Indeed it is much easier to bleed - the
> old system was a bit of a pain at initial fill with airlocks.
open vented system (ie the store will not cope with pressure).
> I'll dig out the installation instructions.
>
--
Tim Watts
Dec 30, 2012, 7:53:57 AM12/30/12
to
Dave Liquorice wrote:
> On Sun, 30 Dec 2012 01:09:15 +0000, Tim Watts wrote:
>
>> Am at any risk of lack of head possibly leading to pump over problems?
>
> Think about the static water levels in the tank and pipes. The water
> level in the open vent pipe will be the same as that in the tank. You
> only need enough "pressure" to lift the water up to the base of the loop
> over the top of the tank to get pump over. The head doesn't really come
> into it.
OK. So are there some good general rules about this somewhere? Is it more
> On Sun, 30 Dec 2012 01:09:15 +0000, Tim Watts wrote:
>
>> Am at any risk of lack of head possibly leading to pump over problems?
>
> Think about the static water levels in the tank and pipes. The water
> level in the open vent pipe will be the same as that in the tank. You
> only need enough "pressure" to lift the water up to the base of the loop
> over the top of the tank to get pump over. The head doesn't really come
> into it.
usual to put pumps in flow or return pipes?
>> Second question - is it usual these days to have a seperate feed to CH
>> from the header tank and a seperate vent pipe hanging over the top - or
>> a single feed in 22mm - with downward slopes of course, all the way?
>
> Presumably the water in the thermal store is the primary water shared by
> the boiler, CH and DHW heat exchanger.
> I'd fit the feed and expansion as
> per normal to the tank via their own tappings if possible. ie feed to the
> bottom and expansion always rising from the very top of the cyclinder. A
> single pipe would have to be connected at the very top to vent any
> air/gases evolved in the store. When the store cools it'll draw
> relatively cold water from the header tank into the top of the store. It
> won't stay there, being denser, it will sink setting up circulation that
> will tend to destroy any stratification in the store.
plate exchanger - this will also be the vent. Radiator taps are at 1/3 - 2/3
heights (as return water is still warm). HW Plate exchanger return will be
at the bottom as I expect fairly cold return temps.
> Note that the feed and vent pipes are now not in any pumped circuit so
> pump over (or draw down) might not be such a big issue but I'd still make
> the vent loop over the top of the haeder a "sensible height".
>
will experience suction rather than +ve pressure. The head at the point will
be about 2m.
I could tak ethe exapnsion pipe for a little circuitous route to the top of
the dormer and back down (another metre gain) if that is likely to help?
Cheers - and thanks for the thoughts :)
Dave Liquorice
Dec 30, 2012, 9:15:40 AM12/30/12
to
On Sun, 30 Dec 2012 12:53:57 +0000, Tim Watts wrote:
>> Think about the static water levels in the tank and pipes. The water
>> level in the open vent pipe will be the same as that in the tank. You
>> only need enough "pressure" to lift the water up to the base of the
>> loop over the top of the tank to get pump over.
>
>> Think about the static water levels in the tank and pipes. The water
>> level in the open vent pipe will be the same as that in the tank. You
>> only need enough "pressure" to lift the water up to the base of the
>> loop over the top of the tank to get pump over.
>
> OK. So are there some good general rules about this somewhere?
There are but I can't remember 'em well off the top of my head. Have a
dig about on't web but no doubt you'll finmd lots of conflicting
information...
> Is it more usual to put pumps in flow or return pipes?
pressure" thus might have a tendancy to draw air in
>> I'd fit the feed and expansion as per normal to the tank via their own
>> tappings if possible. ie feed to the bottom and expansion always
>> rising from the very top of the cyclinder. A single pipe would have to
>> be connected at the very top to vent any air/gases evolved in the
>> store. When the store cools it'll draw relatively cold water from the
>> header tank into the top of the store. It won't stay there, being
>> denser, it will sink setting up circulation that will tend to destroy
>> any stratification in the store.
>
> This is a good point - thank you. I have a top centre tap for feeding
> the HW plate exchanger - this will also be the vent.
rising at say 15 degrees or so. No "nice neat" horizontal section to take
it from the center of the tank to the edge and up the wall to the header.
> Radiator taps are at 1/3 - 2/3 heights (as return water is still warm).
the DHW exchanger.
> HW Plate exchanger return will be at the bottom as I expect fairly cold
> return temps.
return temps? I assume there will be a thermostatic mixer valve on the
actual HW side to limit the HW temp to something sensible. The store
should be operating at 70 to 80C which is far to hot for taps...
Thermo-dynamics say return temp will always be higher than the temp of
the actual HW leaving the plate exchanger, so it may only be 10 or 20
degrees lower than the top of the tank.
The 300l thermal store here has tapings at inches from the store base,
solar coil 4 to 10, CH 13 to 37, HW 42 to 64, oil boiler 13 to 64 and
wood burner 4 to 55. So the CH cannot cool the half of the store used by
the HW. The solar heats as much of the tank as it can with maximum
cooling to that circuit. The wood burner heats most of the volume of the
store, with a bit of safety margin should it get things to boiling point.
The oil boiler only heats the section of tank used, the thermostat for
boiler control is a few inches below the 37 inch level(*).
> Almost - the top pipe will be in the pumped circuit for the HW plate -
> but will experience suction rather than +ve pressure. The head at the
> point will be about 2m.
>
> top of the dormer and back down (another metre gain) if that is likely
> to help?
Not for draw down when the DHW pump starts. What happens when the CH or
> to help?
boiler pumps start is another matter they might lower or raise the
pressure in the store. I'd expect them to lower it if in the flow and
raise it if in the return, assuming they are close to the store...
I think you'd be OK with 18" or so from highest water level in header
tank to the bottom of the vent loop, but don't quote me. B-)
(*) Don't fit it above the the top CH tapping our you'll wonder why the
rads are cool, the heating calling for heat but the boiler not firing...
--
Cheers
Dave.
Dave Plowman (News)
Dec 30, 2012, 9:59:19 AM12/30/12
to
In article <5lp5r9-...@squidward.local.dionic.net>,
> > Can I ask why you don't want a pressurized system? I'm a bit of a
> > dinosaur about such things but happily changed my system when fitting
> > the Viessmann, and have no regrets. Indeed it is much easier to bleed
> > - the old system was a bit of a pain at initial fill with airlocks.
> Because I already have the thermal store cylinder and my design is for
> an open vented system (ie the store will not cope with pressure).
Is there a restriction on how much head you can have to it also? There
can't be *that* much difference between the two.
--
*When cheese gets its picture taken, what does it say? *
Tim Watts <tw+u...@dionic.net> wrote:
> > I've got a Viessmann, and that's why I asked, as I'm pretty sure it
> > said no open header tank. But it's not a combi but a 'system' boiler
> > so comes with internal pressure vessel anyway.
> There's another range that is fine for open vented.
Ah - right.
> > I've got a Viessmann, and that's why I asked, as I'm pretty sure it
> > said no open header tank. But it's not a combi but a 'system' boiler
> > so comes with internal pressure vessel anyway.
> There's another range that is fine for open vented.
> > Can I ask why you don't want a pressurized system? I'm a bit of a
> > dinosaur about such things but happily changed my system when fitting
> > the Viessmann, and have no regrets. Indeed it is much easier to bleed
> > - the old system was a bit of a pain at initial fill with airlocks.
> Because I already have the thermal store cylinder and my design is for
> an open vented system (ie the store will not cope with pressure).
can't be *that* much difference between the two.
--
*When cheese gets its picture taken, what does it say? *
John Rumm
Dec 30, 2012, 2:08:57 PM12/30/12
to
On 30/12/2012 01:09, Tim Watts wrote:
> Just surveying positions for the header tank...
>
> (Please don't say sealed system - I am committed to a vented system)
>
> Now, most of the CH is downstairs and I have a nice position for the header
> tank such that the base (and thus water head) is 150cm *above* the ground
> floor ceiling, and about 130cm above the upstairs floor.
That sounds similar to the system that I had until recently.
> Just surveying positions for the header tank...
>
> (Please don't say sealed system - I am committed to a vented system)
>
> Now, most of the CH is downstairs and I have a nice position for the header
> tank such that the base (and thus water head) is 150cm *above* the ground
> floor ceiling, and about 130cm above the upstairs floor.
> The pumps[1] will be 30-40 cm below the ceiling
to the zone valves. (the feed dropped from the ceiling - it travelled up
from the boiler at the back of the house into the eves roof void = along
across then down)
> [1] Boiler circuit, rads, HW plate exchanger all have sperate pumps
> into/from a thermal store. The header tank feeds the thermal store -
> basically just a 150 litre copper cylinder with extra tappings.
> Am at any risk of lack of head possibly leading to pump over problems?
then tee the F&E pipe into that close to where it joins the main
circuit, then you should eliminate the main problems of sucking air into
the system, and pump over.
> There will probbaly be 2 rads upstairs, short profile, so tops > 50cm under
> the bask of the header tank.
have any inversions then you should be able to avoid air locks.
> Second question - is it usual these days to have a seperate feed to CH from
> the header tank and a seperate vent pipe hanging over the top - or a single
> feed in 22mm - with downward slopes of course, all the way?
(stick a tall loop in it if you can to make the differential pressure
required to pump over higher)
##
# #
| F&E | #
|_____| #
# #
# #
# #
# #
######
#
############| P -> |######
> I *could* get the tank right up to the ceiling upstairs, but it's a bit
> awkward for other reasons - this is a hipped single story bungalow with
> dormers upstairs. Hanging the tank inside the top of the dormer is tricky.
the tank for servicing etc.
--
Cheers,
John.
/=================================================================\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\=================================================================/
Tim Watts
Dec 30, 2012, 2:45:42 PM12/30/12
to
Hi Dave,
Dave Liquorice wrote:
> On Sun, 30 Dec 2012 12:53:57 +0000, Tim Watts wrote:
>
>>> Think about the static water levels in the tank and pipes. The water
>>> level in the open vent pipe will be the same as that in the tank. You
>>> only need enough "pressure" to lift the water up to the base of the
>>> loop over the top of the tank to get pump over.
>>
>> OK. So are there some good general rules about this somewhere?
>
> There are but I can't remember 'em well off the top of my head. Have a
> dig about on't web but no doubt you'll finmd lots of conflicting
> information...
That's what I assumed. Some things are fairly clear cut and some parts of CH
design are not quite so obvious...
>> Is it more usual to put pumps in flow or return pipes?
>
> The return is cooler but will then tend to make the entire circuit "low
> pressure" thus might have a tendancy to draw air in
True... One of those less clear cut areas - except the boiler loop, where it
is sometimes explicitly stated in the installation diagram.
>>> I'd fit the feed and expansion as per normal to the tank via their own
>>> tappings if possible. ie feed to the bottom and expansion always
>>> rising from the very top of the cyclinder. A single pipe would have to
>>> be connected at the very top to vent any air/gases evolved in the
>>> store. When the store cools it'll draw relatively cold water from the
>>> header tank into the top of the store. It won't stay there, being
>>> denser, it will sink setting up circulation that will tend to destroy
>>> any stratification in the store.
>>
>> This is a good point - thank you. I have a top centre tap for feeding
>> the HW plate exchanger - this will also be the vent.
>
> I thought that might be the case. Just remember to keep the vent pipe
> rising at say 15 degrees or so. No "nice neat" horizontal section to take
> it from the center of the tank to the edge and up the wall to the header.
Yes indeed. As I said, I think I can take that up the top of the dormer side
wall and back down. I'll slope it down from the peak both ways and do it in
22mm (it's not a long run, probably 5-6m all over).
I am wonder as to the merit of one or two 3-port air traps in the rad
circuit - the air port can vent into the same pipe.
The boiler circuit will clear itself into the cylinder - short and simple,
ditto the HW plate circuit.
>> Radiator taps are at 1/3 - 2/3 heights (as return water is still warm).
>
> And leaves you a little bit of hot water at the top of the tank to feed
> the DHW exchanger.
That was a deliberate design choice :) Always need some HW. In fact version
1 of the CH will be HW only and electically heated - my tank contains 3
immersion heater bosses - 1 at about the 1/4 from base level[2] and 2 more
above at different radial entry points.
[2] There's a solar coil (sensible future provision) which prevents the
immersion bosses going lower. If I had solar, this would be correct to leave
a cold zone there for the solar to deal with.
I suspect I *may* get localised boiling if I trie to run all 3 at once - but
if 2 work (6kW) then the 3rd can be regarded as an in-place spare. At this
point I will sign up for and Economy 7 type tariff and charge the store
overnight.
Rads will be added on piecemeal and the boiler will be added last when the
system is debugged.
If I do not use the solar coil, I may add a diverter on so the boiler can
take water from the bottom tapping so it can heta 100% of the store.
The core control logic will be 2 tank stats (2/3rds and 1/3 level) and
relays and a time clock. Later I hope to add some computer control that will
override the relays and provide finer control and weather compensation. If
the computer fails, it can be turned off and the relays will provide a
basically sane operation. I love computers, but I do not trust them!
>> HW Plate exchanger return will be at the bottom as I expect fairly cold
>> return temps.
>
> Might be, does the plate exchanger data sheet have anything to say about
> return temps? I assume there will be a thermostatic mixer valve on the
> actual HW side to limit the HW temp to something sensible.
I'm going to try the 3 port mixer on the *feed* to the plateX - it may or
may not work well. If it works, it will reduce stirring in the cylinder
considerably. I'll lash this in plastic and see how it behaves before I
commit to copper.
> The store
> should be operating at 70 to 80C which is far to hot for taps...
> Thermo-dynamics say return temp will always be higher than the temp of
> the actual HW leaving the plate exchanger, so it may only be 10 or 20
> degrees lower than the top of the tank.
Yes - and I will have mixer valves on the rad circuit. I may also have a
lower temp UFH circuit too for 2 locations.
> The 300l thermal store here has tapings at inches from the store base,
> solar coil 4 to 10, CH 13 to 37, HW 42 to 64, oil boiler 13 to 64 and
> wood burner 4 to 55. So the CH cannot cool the half of the store used by
> the HW. The solar heats as much of the tank as it can with maximum
> cooling to that circuit. The wood burner heats most of the volume of the
> store, with a bit of safety margin should it get things to boiling point.
> The oil boiler only heats the section of tank used, the thermostat for
> boiler control is a few inches below the 37 inch level(*).
Sounds nice. I would have liked to have had a solid fuel input too - but I
believed (maybe incorrectly) that I would need a gravity circuit between it
and the stove - and I can;t do that as the cyclinder needs to be on the
ground floor. I will see how those tapping heights compare to mine. From
memory I have a 150l tank (biggest copper cylinder I could find for a sane
price). I have one base tapping, one top, 2 at 1/3 and 2 at 2/3 for CH and
UFH. And 2 coil connections.
>> Almost - the top pipe will be in the pumped circuit for the HW plate -
>> but will experience suction rather than +ve pressure. The head at the
>> point will be about 2m.
>>
>> I could take the expansion pipe for a little circuitous route to the
>> top of the dormer and back down (another metre gain) if that is likely
>> to help?
>
> Not for draw down when the DHW pump starts. What happens when the CH or
> boiler pumps start is another matter they might lower or raise the
> pressure in the store. I'd expect them to lower it if in the flow and
> raise it if in the return, assuming they are close to the store...
It's probably a safer bet - unless I test it in plastic hose first so I can
see exactly what it does.
> I think you'd be OK with 18" or so from highest water level in header
> tank to the bottom of the vent loop, but don't quote me. B-)
Thanks - useful number - that would require my circuitous route as the
header tank will be up against the roof. Yes I did oversize my header tank
to cope with the expansion of about 200l of water! I got a fibreglass tank
rated for 100C as it may see very high temperatures and it will be over my
daughters bed! (Rembering that unfortunate case of the baby being killed by
a boiling over system). There's not much chance of my system boiling as I
would need failures in the tank stat, the immersion stat and the immersion
safety thermal fuse. But better safe than sorry...
> (*) Don't fit it above the the top CH tapping our you'll wonder why the
> rads are cool, the heating calling for heat but the boiler not firing...
>
I *think* (hope) I placed my 2 stat probe tubes at sensible heights!
"History will be kind to me for I intend to write it."
Dave Liquorice wrote:
> On Sun, 30 Dec 2012 12:53:57 +0000, Tim Watts wrote:
>
>>> Think about the static water levels in the tank and pipes. The water
>>> level in the open vent pipe will be the same as that in the tank. You
>>> only need enough "pressure" to lift the water up to the base of the
>>> loop over the top of the tank to get pump over.
>>
>> OK. So are there some good general rules about this somewhere?
>
> There are but I can't remember 'em well off the top of my head. Have a
> dig about on't web but no doubt you'll finmd lots of conflicting
> information...
design are not quite so obvious...
>> Is it more usual to put pumps in flow or return pipes?
>
> The return is cooler but will then tend to make the entire circuit "low
> pressure" thus might have a tendancy to draw air in
is sometimes explicitly stated in the installation diagram.
>>> I'd fit the feed and expansion as per normal to the tank via their own
>>> tappings if possible. ie feed to the bottom and expansion always
>>> rising from the very top of the cyclinder. A single pipe would have to
>>> be connected at the very top to vent any air/gases evolved in the
>>> store. When the store cools it'll draw relatively cold water from the
>>> header tank into the top of the store. It won't stay there, being
>>> denser, it will sink setting up circulation that will tend to destroy
>>> any stratification in the store.
>>
>> This is a good point - thank you. I have a top centre tap for feeding
>> the HW plate exchanger - this will also be the vent.
>
> I thought that might be the case. Just remember to keep the vent pipe
> rising at say 15 degrees or so. No "nice neat" horizontal section to take
> it from the center of the tank to the edge and up the wall to the header.
wall and back down. I'll slope it down from the peak both ways and do it in
22mm (it's not a long run, probably 5-6m all over).
I am wonder as to the merit of one or two 3-port air traps in the rad
circuit - the air port can vent into the same pipe.
The boiler circuit will clear itself into the cylinder - short and simple,
ditto the HW plate circuit.
>> Radiator taps are at 1/3 - 2/3 heights (as return water is still warm).
>
> And leaves you a little bit of hot water at the top of the tank to feed
> the DHW exchanger.
1 of the CH will be HW only and electically heated - my tank contains 3
immersion heater bosses - 1 at about the 1/4 from base level[2] and 2 more
above at different radial entry points.
[2] There's a solar coil (sensible future provision) which prevents the
immersion bosses going lower. If I had solar, this would be correct to leave
a cold zone there for the solar to deal with.
I suspect I *may* get localised boiling if I trie to run all 3 at once - but
if 2 work (6kW) then the 3rd can be regarded as an in-place spare. At this
point I will sign up for and Economy 7 type tariff and charge the store
overnight.
Rads will be added on piecemeal and the boiler will be added last when the
system is debugged.
If I do not use the solar coil, I may add a diverter on so the boiler can
take water from the bottom tapping so it can heta 100% of the store.
The core control logic will be 2 tank stats (2/3rds and 1/3 level) and
relays and a time clock. Later I hope to add some computer control that will
override the relays and provide finer control and weather compensation. If
the computer fails, it can be turned off and the relays will provide a
basically sane operation. I love computers, but I do not trust them!
>> HW Plate exchanger return will be at the bottom as I expect fairly cold
>> return temps.
>
> Might be, does the plate exchanger data sheet have anything to say about
> return temps? I assume there will be a thermostatic mixer valve on the
> actual HW side to limit the HW temp to something sensible.
may not work well. If it works, it will reduce stirring in the cylinder
considerably. I'll lash this in plastic and see how it behaves before I
commit to copper.
> The store
> should be operating at 70 to 80C which is far to hot for taps...
> Thermo-dynamics say return temp will always be higher than the temp of
> the actual HW leaving the plate exchanger, so it may only be 10 or 20
> degrees lower than the top of the tank.
lower temp UFH circuit too for 2 locations.
> The 300l thermal store here has tapings at inches from the store base,
> solar coil 4 to 10, CH 13 to 37, HW 42 to 64, oil boiler 13 to 64 and
> wood burner 4 to 55. So the CH cannot cool the half of the store used by
> the HW. The solar heats as much of the tank as it can with maximum
> cooling to that circuit. The wood burner heats most of the volume of the
> store, with a bit of safety margin should it get things to boiling point.
> The oil boiler only heats the section of tank used, the thermostat for
> boiler control is a few inches below the 37 inch level(*).
believed (maybe incorrectly) that I would need a gravity circuit between it
and the stove - and I can;t do that as the cyclinder needs to be on the
ground floor. I will see how those tapping heights compare to mine. From
memory I have a 150l tank (biggest copper cylinder I could find for a sane
price). I have one base tapping, one top, 2 at 1/3 and 2 at 2/3 for CH and
UFH. And 2 coil connections.
>> Almost - the top pipe will be in the pumped circuit for the HW plate -
>> but will experience suction rather than +ve pressure. The head at the
>> point will be about 2m.
>>
>> I could take the expansion pipe for a little circuitous route to the
>> top of the dormer and back down (another metre gain) if that is likely
>> to help?
>
> Not for draw down when the DHW pump starts. What happens when the CH or
> boiler pumps start is another matter they might lower or raise the
> pressure in the store. I'd expect them to lower it if in the flow and
> raise it if in the return, assuming they are close to the store...
see exactly what it does.
> I think you'd be OK with 18" or so from highest water level in header
> tank to the bottom of the vent loop, but don't quote me. B-)
header tank will be up against the roof. Yes I did oversize my header tank
to cope with the expansion of about 200l of water! I got a fibreglass tank
rated for 100C as it may see very high temperatures and it will be over my
daughters bed! (Rembering that unfortunate case of the baby being killed by
a boiling over system). There's not much chance of my system boiling as I
would need failures in the tank stat, the immersion stat and the immersion
safety thermal fuse. But better safe than sorry...
> (*) Don't fit it above the the top CH tapping our you'll wonder why the
> rads are cool, the heating calling for heat but the boiler not firing...
>
"History will be kind to me for I intend to write it."
Tim Watts
Dec 30, 2012, 4:18:12 PM12/30/12
to
Hi John,
John Rumm wrote:
> On 30/12/2012 01:09, Tim Watts wrote:
>> Just surveying positions for the header tank...
>>
>> (Please don't say sealed system - I am committed to a vented system)
>>
>> Now, most of the CH is downstairs and I have a nice position for the
>> header tank such that the base (and thus water head) is 150cm *above* the
>> ground floor ceiling, and about 130cm above the upstairs floor.
>
> That sounds similar to the system that I had until recently.
>
>> The pumps[1] will be 30-40 cm below the ceiling
>
> Ours was probably 2m below the ceiling in the feed pipe before the split
> to the zone valves. (the feed dropped from the ceiling - it travelled up
> from the boiler at the back of the house into the eves roof void = along
> across then down)
>
>> [1] Boiler circuit, rads, HW plate exchanger all have sperate pumps
>> into/from a thermal store. The header tank feeds the thermal store -
>> basically just a 150 litre copper cylinder with extra tappings.
>
> No indirect coil I take it?
No...
>> Am at any risk of lack of head possibly leading to pump over problems?
>
> If you take your expansion pipe from the suction side of the pump, and
> then tee the F&E pipe into that close to where it joins the main
> circuit, then you should eliminate the main problems of sucking air into
> the system, and pump over.
That's an interesting design - I never thought of that.
>> There will probbaly be 2 rads upstairs, short profile, so tops > 50cm
>> under the bask of the header tank.
>
> They will probably still be a PITA to bleed, but if your pipe runs don't
> have any inversions then you should be able to avoid air locks.
No inversions on the 1st floor. There will be loads of inversions on the GF
as it's solid so all drops from above (essentially there is one radiator bus
circuit and *everything* comes off it.
>> Second question - is it usual these days to have a seperate feed to CH
>> from the header tank and a seperate vent pipe hanging over the top - or a
>> single feed in 22mm - with downward slopes of course, all the way?
>
> You need the vent pipe hanging over the top as a safety precaution
> (stick a tall loop in it if you can to make the differential pressure
> required to pump over higher)
>
> ##
> # #
> | F&E | #
> |_____| #
> # #
> # #
> # #
> # #
> ######
> #
> ############| P -> |######
OK - thanks :)
>
>> I *could* get the tank right up to the ceiling upstairs, but it's a bit
>> awkward for other reasons - this is a hipped single story bungalow with
>> dormers upstairs. Hanging the tank inside the top of the dormer is
>> tricky.
>
> Higher it is, easier to bleed the rads etc, although harder to get at
> the tank for servicing etc.
This position is at the top of the stairs - inside a cupboard I will make in
the wall under the roof space (stairs come up right at the apex of hips and
ridge to one side of the dormer centre line). It's actuall quite convenient
from every aspect.
Sounds like it is a goer then...
Cheers,
John Rumm wrote:
> On 30/12/2012 01:09, Tim Watts wrote:
>> Just surveying positions for the header tank...
>>
>> (Please don't say sealed system - I am committed to a vented system)
>>
>> Now, most of the CH is downstairs and I have a nice position for the
>> header tank such that the base (and thus water head) is 150cm *above* the
>> ground floor ceiling, and about 130cm above the upstairs floor.
>
> That sounds similar to the system that I had until recently.
>
>> The pumps[1] will be 30-40 cm below the ceiling
>
> Ours was probably 2m below the ceiling in the feed pipe before the split
> to the zone valves. (the feed dropped from the ceiling - it travelled up
> from the boiler at the back of the house into the eves roof void = along
> across then down)
>
>> [1] Boiler circuit, rads, HW plate exchanger all have sperate pumps
>> into/from a thermal store. The header tank feeds the thermal store -
>> basically just a 150 litre copper cylinder with extra tappings.
>
> No indirect coil I take it?
>> Am at any risk of lack of head possibly leading to pump over problems?
>
> If you take your expansion pipe from the suction side of the pump, and
> then tee the F&E pipe into that close to where it joins the main
> circuit, then you should eliminate the main problems of sucking air into
> the system, and pump over.
>> There will probbaly be 2 rads upstairs, short profile, so tops > 50cm
>> under the bask of the header tank.
>
> They will probably still be a PITA to bleed, but if your pipe runs don't
> have any inversions then you should be able to avoid air locks.
as it's solid so all drops from above (essentially there is one radiator bus
circuit and *everything* comes off it.
>> Second question - is it usual these days to have a seperate feed to CH
>> from the header tank and a seperate vent pipe hanging over the top - or a
>> single feed in 22mm - with downward slopes of course, all the way?
>
> You need the vent pipe hanging over the top as a safety precaution
> (stick a tall loop in it if you can to make the differential pressure
> required to pump over higher)
>
> ##
> # #
> | F&E | #
> |_____| #
> # #
> # #
> # #
> # #
> ######
> #
> ############| P -> |######
>
>> I *could* get the tank right up to the ceiling upstairs, but it's a bit
>> awkward for other reasons - this is a hipped single story bungalow with
>> dormers upstairs. Hanging the tank inside the top of the dormer is
>> tricky.
>
> Higher it is, easier to bleed the rads etc, although harder to get at
> the tank for servicing etc.
the wall under the roof space (stairs come up right at the apex of hips and
ridge to one side of the dormer centre line). It's actuall quite convenient
from every aspect.
Sounds like it is a goer then...
Cheers,
Onetap
Dec 30, 2012, 5:56:37 PM12/30/12
to Dave Liquorice
On Sunday, December 30, 2012 1:09:15 AM UTC, Tim Watts wrote:
> (Please don't say sealed system - I am committed to a vented system)
Why so?
> (Please don't say sealed system - I am committed to a vented system)
Open vents are fine, they work, I had one for a long time.
I'd go for a sealed system if possible, they're better.
On Sunday, December 30, 2012 11:37:42 AM UTC, Dave Liquorice wrote:
> Think about the static water levels in the tank and pipes. The water
>
> level in the open vent pipe will be the same as that in the tank. You
>
Ain't necessarily so.
What is critical is the location of the cold feed connection and the open vent connection.
There is a loss of pressure/head as water flows along a pipe, or through a fitting due to the frictional resistance.
There is a loss of energy & this appears as a loss of pressure (Bernouilli, etc.).
If the difference in the head at the two connections is high enough when the pump is running,
then water or air will flow down one pipe and up the other.
In the olden days, the cold feed was on the return side of the boiler and the open vent on the flow side.
This was when boilers were cast iron with big waterways, little hydraulic resistance (and so little head loss across return & flow)
and solid fuel boilers did boil occasionally.
Steam had to escape up the flow pipe, make-up water had to enter with the return water.
Then someone invented efficiency, cast-iron boilers were no longer made and waterways got much smaller.
A 'traditional' arrangement of OV & CF ensured pumping over, especially where a cast-iron boiler was replaced.
People complained them new-fangled high-efficiency boilers didn't last as long; funny, that.
Dom Ostrowski
Dec 30, 2012, 6:17:16 PM12/30/12
to
On 30/12/12 01:09, Tim Watts wrote:
> Just surveying positions for the header tank...
I think most of your questions have been answered already, and you seem
> Just surveying positions for the header tank...
to have most of your issues in hand anyway.
Your system is very similar to mine, apart from mine uses an in-tank
extra-long coil in place of the heat exchanger for hot water (so stored
HW would still be available during a power-cut).
Mine has been absolutely trouble free, behaves ideally, and has been
returning excellent economy (no gas here, so modern oil-fired
pressure-jet Rayburn).
Such tips as I can give you:
Header tank - as high as possible - makes filling/bleeding easier
(chapel conversion here, so about an 8m head overall, and 2.5m to the
next highest point - the upstairs rads).
Use a conventional vent pipe in addition to the feed pipe to the header
tank (there's a formula for calculating how high above the tank the vent
pipe should rise). The vent pipe is a critical safety feature - imagine
if the feed pipe froze or became blocked by debris.
Pumps for rads pump from the store, to the rads (2 circuits here).
Pump for boiler pumps from store to boiler.
Those CH circulation filters (like the Fernox TF1 - recommended) -
they're good things, and a convenient place to add water treatment.
Add lots of drain-off points.
Tim Watts
Dec 30, 2012, 6:51:59 PM12/30/12
to
Hi Dom,
Dom Ostrowski wrote:
> On 30/12/12 01:09, Tim Watts wrote:
>> Just surveying positions for the header tank...
>
>
> I think most of your questions have been answered already, and you seem
> to have most of your issues in hand anyway.
>
> Your system is very similar to mine, apart from mine uses an in-tank
> extra-long coil in place of the heat exchanger for hot water (so stored
> HW would still be available during a power-cut).
Good move. I'm going to run the boiler/control/pump feeds through a single
13A plug fed by a socket on it's own RCBO. Next to this socket will be a
second socket on a cable to an external mail socket (probably "commando
style) that I can plug a generator in if necessary. I have other reasons to
plan this in - such as providing feeds to core computing equipment in case
our supply goes out for more than the odd hour (as happens a couple of times
a year).
> Mine has been absolutely trouble free, behaves ideally, and has been
> returning excellent economy (no gas here, so modern oil-fired
> pressure-jet Rayburn).
>
> Such tips as I can give you:
>
> Header tank - as high as possible - makes filling/bleeding easier
> (chapel conversion here, so about an 8m head overall, and 2.5m to the
> next highest point - the upstairs rads).
I'm guessing you have a 2 storey house and the tank's in the loft? My first
floor is the loft, so I'm a bit limited.
> Use a conventional vent pipe in addition to the feed pipe to the header
> tank (there's a formula for calculating how high above the tank the vent
> pipe should rise). The vent pipe is a critical safety feature - imagine
> if the feed pipe froze or became blocked by debris.
Indeed :) Good point on debris.
> Pumps for rads pump from the store, to the rads (2 circuits here).
> Pump for boiler pumps from store to boiler.
OK...
> Those CH circulation filters (like the Fernox TF1 - recommended) -
> they're good things, and a convenient place to add water treatment.
That is interesting... Cheers.
> Add lots of drain-off points.
I am actually going to rig a 40mm drain pipe (with shallow trap and screw on
cap) right under the primary drain cock on the cylinder - so a 6" bit of
hose can dump direct to drain. Only because the main drains for the kitchen
and bath converge a few feet away...
Every ground floor rad will need drain cocks as well as pipes must drop down
from the ceiling.
Thanks for all that -
Cheers
Dom Ostrowski wrote:
> On 30/12/12 01:09, Tim Watts wrote:
>> Just surveying positions for the header tank...
>
>
> I think most of your questions have been answered already, and you seem
> to have most of your issues in hand anyway.
>
> Your system is very similar to mine, apart from mine uses an in-tank
> extra-long coil in place of the heat exchanger for hot water (so stored
> HW would still be available during a power-cut).
13A plug fed by a socket on it's own RCBO. Next to this socket will be a
second socket on a cable to an external mail socket (probably "commando
style) that I can plug a generator in if necessary. I have other reasons to
plan this in - such as providing feeds to core computing equipment in case
our supply goes out for more than the odd hour (as happens a couple of times
a year).
> Mine has been absolutely trouble free, behaves ideally, and has been
> returning excellent economy (no gas here, so modern oil-fired
> pressure-jet Rayburn).
>
> Such tips as I can give you:
>
> Header tank - as high as possible - makes filling/bleeding easier
> (chapel conversion here, so about an 8m head overall, and 2.5m to the
> next highest point - the upstairs rads).
floor is the loft, so I'm a bit limited.
> Use a conventional vent pipe in addition to the feed pipe to the header
> tank (there's a formula for calculating how high above the tank the vent
> pipe should rise). The vent pipe is a critical safety feature - imagine
> if the feed pipe froze or became blocked by debris.
> Pumps for rads pump from the store, to the rads (2 circuits here).
> Pump for boiler pumps from store to boiler.
> Those CH circulation filters (like the Fernox TF1 - recommended) -
> they're good things, and a convenient place to add water treatment.
> Add lots of drain-off points.
cap) right under the primary drain cock on the cylinder - so a 6" bit of
hose can dump direct to drain. Only because the main drains for the kitchen
and bath converge a few feet away...
Every ground floor rad will need drain cocks as well as pipes must drop down
from the ceiling.
Thanks for all that -
Cheers
0 new messages