Central heating pump bypass
Dave Sawdon
why it's there. The 3-way valve has a mid-position so my understanding
is that a bypass isn't needed. I've shut off the gate valve without any
obvious effects - any comments?
Dave S
Phil Addison
This was raised last October and I offered the following but got no
feedback. Better luck this time?
On Thu Oct 26 19:18:26 2000 Phil Addison wrote in message "CH
Balancing" news:aGn4OZk+FfzABz...@4ax.com...
"
I'm no authority, so hopefully someone will comment on the following,
but I recall that a boiler I had said the by-pass should be adjusted so
that there was always some specified minimum flow rate through the
boiler.
I assume this is because in the event of all the rads getting turned off
there might be (a) a delay before the flame goes out, and (b) the
combustion chamber (at least the old cast-iron ones) could still hold
enough heat to boil the contents.
According to this theory, the bypass circuit should not only have a
certain flow rate, but also hold enough water to absorb this surplus
heating. I suppose the dissipation from the bypass and primary pipework
is fairly negligible. Of course the boiler maker never hinted at how to
set up the gate valve to achieve the specified flow rate.
Given that my theory is correct, one could set up the bypass with all
the rads (and hot water cylinder) shut off. In this case the full pump
pressure is applied to the bypass so the flow rate should be achieved
with quite a small opening of the valve. My guess is that you start with
the valve open, then and with the boiler firing you close the valve
until the boiler starts knocking and then open it up a bit. When the
boiler overheat stat cuts out you just have to let things cool down and
have another go. The final result should be that with the c/h running
flat out, you can shut down the rad valves and have the boiler run up
till its overheat stat cycles on and off without nasty noises coming
from the boiler.
When the rads are on, as normal, the pump pressure across the bypass
will be much reduced and (hopefully) not too much water is then
recirculated round the bypass.
"
You can get an "automatic by-pass valve" that senses when the rads close
down and then opens up the bypass.
Phil Addison
Ledswinger
news:Rwx4OjJFGhmSFd...@4ax.com...
> On Wed, 31 Jan 2001 09:40:16 +0000, in uk.d-i-y Dave Sawdon
> <nom...@hursley.ibm.com> wrote:
> >My CH pump has a bypass pipe (with a gate valve) but I don't understand
> >why it's there. The 3-way valve has a mid-position so my understanding
> >is that a bypass isn't needed. I've shut off the gate valve without any
> >obvious effects - any comments?
>
> This was raised last October and I offered the following but got no
> feedback. Better luck this time?
>
The bypass is simply there to protect the pump if all automatic valves are
shut *or the system is blocked for any other reason*. This fault condition
would cause the pump to fail mechanically, leak or overheat as it endeavours
to pump against a near infinite head. In such fault situations, the bypass is
not normally intended to act as a heat sink, since the boiler has (or should
have) fail safe design and a water thermostat to avoid problems caused by loss
of circulation.
Depending on the design of a three way valve, it might be capable of failing
between settings and in effect closing - although that depends on the design.
But a bypass is good practice in any system, given that there are other,
albeit rare, reasons why circulation could fail.
The bypass should always be left partly open - it doesn't materially reduce
circulation in a healthy, well balanced system, and if things go pear shaped
should prevent one problem leading to another. By having the gate valve
partly closed (about half or two thirds) you reduce the marginal head loss in
normal operation, and in a fault situation give some load for the pump to work
against (as opposed to a simple short circuit, which could give rise to out of
tolerance flow velocities).
Led
Phil Addison
<Ledsw...@bigfoot.charlie.oscar.mike> wrote:
>Phil Addison <phi...@bigfoot.com> wrote in message
>news:Rwx4OjJFGhmSFd...@4ax.com...
>> On Wed, 31 Jan 2001 09:40:16 +0000, in uk.d-i-y Dave Sawdon
>> <nom...@hursley.ibm.com> wrote:
>> >My CH pump has a bypass pipe (with a gate valve) but I don't understand
>> >why it's there. The 3-way valve has a mid-position so my understanding
>> >is that a bypass isn't needed. I've shut off the gate valve without any
>> >obvious effects - any comments?
>>
>> This was raised last October and I offered the following but got no
>> feedback. Better luck this time?
>>
>
>The bypass is simply there to protect the pump if all automatic valves are
>shut *or the system is blocked for any other reason*. This fault condition
>would cause the pump to fail mechanically, leak or overheat as it endeavours
>to pump against a near infinite head. In such fault situations, the bypass is
>not normally intended to act as a heat sink, since the boiler has (or should
>have) fail safe design and a water thermostat to avoid problems caused by loss
>of circulation.
Thanks for that, Led. It certainly sheds some new light on the subject.
I take the point but need to look at some pump sheets to be convinced. I
await other comments with interest.
>But a bypass is good practice in any system, given that there are other,
>albeit rare, reasons why circulation could fail.
You mean as in 'getting blocked, with the pump still running'?
>The bypass should always be left partly open - it doesn't materially reduce
>circulation in a healthy, well balanced system, and if things go pear shaped
>should prevent one problem leading to another.
I dislike 'rules of thumb' such as this. I want to know how to set
things properly according to some theory of operation.
>By having the gate valve
>partly closed (about half or two thirds) you reduce the marginal head loss in
>normal operation,
With the bypass being right between the main flow and return (i.e. right
across the pump) there could be quite a large head loss, which for some
folk could mean inadequate flow to the rest of the system.
>and in a fault situation give some load for the pump to work
>against (as opposed to a simple short circuit, which could give rise to out of
>tolerance flow velocities).
I don't follow that.
Phil Addison
Ed Sirett
Phil Addison wrote in message <=F94Ovi9j6iOA4DaBHnW=dtT...@4ax.com>...
Firstly, Dave Sawdon should put the bypass valve back to its original
setting. I think the issue is not so much that the boiler might boil if
there is insufficient flow, but that the following problems might occur
when the flow rate through the boiler gets too low:
1) (An increase in) Kettling/Banging/Wheezing noises.
2) Possible over heating causing lock-outs onsome boilers.
3) Localized over-heating and even damage in some low water content heat
exchangers.
>Thanks for that, Led. It certainly sheds some new light on the subject.
>I take the point but need to look at some pump sheets to be convinced.
I
>await other comments with interest.
>
The pump is a centrifugal type, these have a maximum pressure (typically
about 0.5-1.0 bar) when there is now flow through them they pressure
developed by the pump drops off as the through flow inreases.
Ed Sirett
Property Maintainer - North London.
Phil Addison
quote]
On Wed, 31 Jan 2001 23:39:43 -0000, in uk.d-i-y "Ed Sirett"
<e...@makewrite.demon.co.uk> wrote:
>Phil Addison wrote in message <=F94Ovi9j6iOA4DaBHnW=dtT...@4ax.com>...
>
>>Thanks for that, Led. It certainly sheds some new light on the subject.
>
>Firstly, Dave Sawdon should put the bypass valve back to its original
>setting.
Agreed. When experimenting it is always wise to note the original and
return to that if you didn't learn a better option from the experiment.
>I think the issue is not so much that the boiler might boil if
>there is insufficient flow, but that the following problems might occur
>when the flow rate through the boiler gets too low:
>1) (An increase in) Kettling/Banging/Wheezing noises.
>2) Possible over heating causing lock-outs onsome boilers.
>3) Localized over-heating and even damage in some low water content heat
>exchangers.
I'll buy that. It still leaves the question of what is a logical way to
set up the bypass, given that there is no practical way of actually
measuring the flow rate through the boiler. In other words we can't
objectively set the bypass rate to the minimum galls/min that the
manufacturer states, we have to use some other observable effect. Is the
procedure I outlined reasonable, or is there another way?
>The pump is a centrifugal type, these have a maximum pressure (typically
>about 0.5-1.0 bar) when there is no flow through them. The pressure
>developed by the pump drops off as the through flow increases.
Exactly. And the pressure drops to zero (by definition) when there is a
short-circuit across it. The flow rate is then at its maximum, with
water going only round the short-circuit. All the pump data sheets show
this graph, but I don't recall any saying that there is a prohibited
operating region (I still haven't had a chance to check this).
>>I await other comments with interest.
Particularly on procedures for setting the bypass, and to debate the
case of not having a bypass at all.
Phil Addison
Tim
wrote:
>>>I await other comments with interest.
>
>Particularly on procedures for setting the bypass, and to debate the
>case of not having a bypass at all.
>
FWIW, I fitted an automatic differential bypass valve. Fitting one of
these (albeit at about 20 UKP costs IIRC) means it is straightforward
to set correctly, and will not compromise the flow rate throught the
rads under normal conditions.
Although you may not want to replace yours, there might be some useful
information to glean from it's functionality, and set-up procedures?
It senses the pressure difference across flow and return and is
supposed to be fitted within 1m of the boiler.
It is normally closed, but opens up when the pressure difference
exceeds a preset limit, between 0.1 and 0.6 bar IIRC.
This can be done in one of two ways, depending on whether it's acting
as a bypass for a full-TRV system, or merely as a system safety
bypass.
Set-up for simple bypass operation is essentially to find the point on
the dial at which the return connection to the valve starts increasing
in temperature - i.e. as the hotter flow water starts to pass through
the valve - and then to back off half a setting. This would, I
believe, result in a valve which will open if the pressure increases
by 0.05bar.
For full-TRV it is broadly similar but is adjusted to acheive "quiet
operation under all working conditions" - which I interpret as still
quiet (no kettling etc) when all TRV's are shut down.
Don't know if this helps - but I've written it now anyway!
Tim Hardisty.
Ledswinger
>
OT: Phil, you are aware of the impact that Bigfoot's ORBS changes are having
on e-mail? I'm not depending on the Ledswinger@bigfoot address for anything
other than newgroup responses so it isn't a big deal here, but for personal
mail I've had to drop my realname@bigfoot and revert to an ISP address, due to
rejection of perhaps 30% of email people have sent me. Mostly the rejection
affects subscribers of small ISP's, and anyone using hosted corporate sites,
including big names like Marconi, Corus, Rolls Royce.
> I'll buy that. It still leaves the question of what is a logical way to
> set up the bypass, given that there is no practical way of actually
> measuring the flow rate through the boiler. In other words we can't
> objectively set the bypass rate to the minimum galls/min that the
> manufacturer states, we have to use some other observable effect. Is the
> procedure I outlined reasonable, or is there another way?
>
I'm willing to be corrected on the setting issue, but observation, and
consideration of the problem, suggest that it is every bit as "rule of thumb"
as any other plumbing job, and about two thirds shut would be a reasonable
guess since you want to mimic the load of the system in very rough terms.
I've certainly never seen a plumber do anything than give the bypass gate a
good twirl!
> Particularly on procedures for setting the bypass, and to debate the
> case of not having a bypass at all.
>
Well, since the maximum downside of having an open bypass is a slight loss of
head, which in a suitably "designed" system will not impact on system
performance, I don't think that once installed there is any point in either
removing or closing it. Experimenting with our installation indicates that
there are no evident noise effects of having it open or closed under normal
running, nor any perceptible improvements in performance with it closed.
Led
Dave Sawdon
You said "within 1m of the boiler", did you mean "within 1m of the pump"?
I thought the bypass was to ensure a flow through the pump (presumably to
keep it operating at a reasonable back pressure?).
Dave
Tim
Dave,
Yes, you're right. In my case it's a system boiler and the pump is
inside the boiler casing so it's one and the same thing.
Apologies for misleading you/anyone.
The instructions for the Honeywell valve assume the pump and boiler
are co-located IIRC.
FWIW the Baxi Barcelona I have requires a minimum flow through the
heat exchanger to avoid the localised over-heating and potential
damage that Ed referred to.
Tim Hardisty.
Mark Dumbrill
with a three way valve there is always a route for the flow - assuming
that there is at least one radiator that doesn't have a TRV fitted.
If the the bypass is open then a large proportion of the pumped flow
will simply return back to the bolier without going through the
radiators or HW tank; not very efficient.
In principle dead heading the pump doesn't necessarily mean that it
will be damaged. It will simply produce its maximum head at no flow.
Conversly providing little or no resitance will allow it to run to the
end of its curve with maximum flow and very little head - how far along
the curve being dictated by the maximum power of the motor. The only
provision to this is that when dead headed for a long time the pump
motor could overheat since there would be little or no flow around the
motor rotor. I would guess that the motor relies on some internal
circulation for lubricating the bearings and cooling, but then only the
pump manufacturer would know this. Dead heading for a short time, a few
minutes or so, shouldn't matter whatever the requirements.
In the end though the simple answer should be to close the bypass and
make sure that there is at least one radiator without a TRV. Or fit a
pressure dependent bypass valve as suggested elsewhere.
Mark Dumbrill
Sent via Deja.com
http://www.deja.com/
Jim Easterbrook
> FWIW, I fitted an automatic differential bypass valve. Fitting one of
> these (albeit at about 20 UKP costs IIRC) means it is straightforward
> to set correctly, and will not compromise the flow rate throught the
> rads under normal conditions.
I've also fitted one, as I have a low water content boiler and am keen
not to damage it.
> For full-TRV it is broadly similar but is adjusted to acheive "quiet
> operation under all working conditions" - which I interpret as still
> quiet (no kettling etc) when all TRV's are shut down.
I set mine as follows:
1/ set bypass to fully closed and balance system (all TRVs fully open)
2/ adjust bypass until it opens (outlet pipe heats up, and I can hear
it)
3/ back off bypass till it closes
4/ close all TRVs and manual rad valves, ensure boiler temperature
difference is within limits
The problem of ensuring adequate flow rate through the boiler, with no
means of measuring same, has been mentioned elsewhere. If the boiler has
a fixed heat input (i.e. not modulating) then surely the flow rate can
be inferred from the temperature rise across the boiler while it's
firing. Although with all the valves closed, you don't get much time to
measure this before the boiler stat shuts it down. Fortunately I bought
two multimeters with thermocouples.
--
Jim Easterbrook <http://www.easter.mersinet.co.uk/>
Ed Sirett
Mark Dumbrill wrote in message <95ceg3$7fe$1...@nnrp1.deja.com>...
>In article <3A77DD80...@hursley.ibm.com>,
> Dave Sawdon <nom...@hursley.ibm.com> wrote:
>> My CH pump has a bypass pipe (with a gate valve) but I don't
>>understand why it's there. The 3-way valve has a mid-position so my
>>understanding is that a bypass isn't needed. I've shut off the gate
>>valve without any obvious effects - any comments?
>>
the valve moves to CH only (because the HW tank is now fully heated) and
then house is fully warm and ALL the TRV's shutdown you'll have a
blocked circuit.
Toby Mack
Ed Sirett <e...@makewrite.demon.co.uk> wrote in message
news:981067104.2650.0...@news.demon.co.uk...
I'm planning on putting TRVs on all my radiators bar the bathroom and loo
(as I decorate). Would this mean I need to add a bypass (my previous house
had one but this one doesn't) or am I likely to get away without as I will
never completely shut the cct?
Ta,
Toby
Dave Liquorice
> I'm planning on putting TRVs on all my radiators bar the bathroom and loo
> (as I decorate). Would this mean I need to add a bypass (my previous house
> had one but this one doesn't) or am I likely to get away without as I will
> never completely shut the cct?
Fine as theory but can you 100% guarantee that those rads will never ever be
turned off? No you can't, even if you don't do it how about the next owners?
For the price a gate valve and two T's it's not worth it but I guess the
same applies to that solution as has been evidenced by this(?) thread. Best
do it properly with an automatic bypass valve.
--
Cheers new...@howhill.com
Dave. Remove "spam" for valid email.
Dave Sawdon
I may well be wrong here but the system SEEMS to work like this:
- The boiler is simply controlled by its internal stat, if the water is too
cold the burner fires until the stat trips and the burner turns off.
- The valve motors to the position demanded by the CH and HW.
- The pump only runs when the CH or HW stats demand heat AND the valve is in
the correct position.
If this is correct there would not be a problem even with all TLVs shut. In
practise I have several rads without TLVs so I still don't understand what
the problem would be.
Having looked again at the position of the bypass pipe I've realised that
it's between the pump outlet (before the 3-way valve) and the primary return
from the HW tank - not directly around the pump as I first thought.
Dave
Phil Addison
<nom...@hursley.ibm.com> wrote:
>Ed,
>
>I may well be wrong here but the system SEEMS to work like this:
>- The boiler is simply controlled by its internal stat, if the water is too
>cold the burner fires until the stat trips and the burner turns off.
Yes
>- The valve motors to the position demanded by the CH and HW.
Yes, under control of programmer. The valves have end switches so that
the boiler only fires once they are fully open.
>- The pump only runs when the CH or HW stats demand heat AND the valve is in
>the correct position.
No. The pump runs whenever the heat is demanded, plus a run-on
afterwards on most systems. The pump keeps running even if the boiler
stat has cut the burner.
Phil Addison
Phil Addison
wrote:
>FWIW, I fitted an automatic differential bypass valve. ..
>It is normally closed, but opens up when the pressure difference
>exceeds a preset limit, between 0.1 and 0.6 bar IIRC.
I'm not clear what this means. Is it adjustable to open somewhere in
that range, and then opens within .05 bar (from below) of the setting?
>This can be done in one of two ways, depending on whether it's acting
>as a bypass for a full-TRV system, or merely as a system safety
>bypass.
>
>Set-up for simple bypass operation is essentially to find the point on
>the dial at which the return connection to the valve starts increasing
>in temperature - i.e. as the hotter flow water starts to pass through
>the valve - and then to back off half a setting. This would, I
>believe, result in a valve which will open if the pressure increases
>by 0.05bar.
>For full-TRV it is broadly similar but is adjusted to acheive "quiet
>operation under all working conditions" - which I interpret as still
>quiet (no kettling etc) when all TRV's are shut down.
Beautiful -the later is just what I suggested . :-)
>Don't know if this helps - but I've written it now anyway!
Yes, thanks Tim, very welcome info.
Phil Addison
Phil Addison
(Jim Easterbrook) wrote:
About automatic by-pass valve...
>I set mine as follows:
>1/ set bypass to fully closed and balance system (all TRVs fully open)
and all control valves open where no TRVs.
>2/ adjust bypass until it opens (outlet pipe heats up, and I can hear
>it)
>3/ back off bypass till it closes
>4/ close all TRVs and manual rad valves, ensure boiler temperature
>difference is within limits
Huh? With all valves shut the boiler will cycle on its overheat stat.
The return will be non-existent (i.e. cold).
>The problem of ensuring adequate flow rate through the boiler, with no
>means of measuring same, has been mentioned elsewhere.
That was me.
> If the boiler has
>a fixed heat input (i.e. not modulating) then surely the flow rate can
>be inferred from the temperature rise across the boiler while it's
>firing.
There is no way of estimating the flow numerically. All you can say is
that there is some, and possibly that it is high enough if the boiler
stat does not cut out.
>Although with all the valves closed, you don't get much time to
>measure this before the boiler stat shuts it down.
No indeed. And no flow to 'estimate' either!!
>Fortunately I bought
>two multimeters with thermocouples.
For balancing I imagine, not for flow-rate.
Phil Addison
Phil Addison
<Ledsw...@bigfoot.charlie.oscar.mike> wrote:
>Phil Addison <phi...@bigfoot.com> wrote in message
>news:hq94Or6MgBfiSF1twu7=MhuG...@4ax.com...
>>
>
>OT: Phil, you are aware of the impact that Bigfoot's ORBS changes are having
>on e-mail? I'm not depending on the Ledswinger@bigfoot address for anything
>other than newgroup responses so it isn't a big deal here, but for personal
>mail I've had to drop my realname@bigfoot and revert to an ISP address, due to
>rejection of perhaps 30% of email people have sent me. Mostly the rejection
>affects subscribers of small ISP's, and anyone using hosted corporate sites,
>including big names like Marconi, Corus, Rolls Royce.
Not to mention Microsoft!! Yes, thanks Led, it's nice and peaceful here.
If anyone is not getting replies from me please post and I'll give you a
direct email address.
>> I'll buy that. It still leaves the question of what is a logical way to
>> set up the bypass, given that there is no practical way of actually
>> measuring the flow rate through the boiler. In other words we can't
>> objectively set the bypass rate to the minimum galls/min that the
>> manufacturer states, we have to use some other observable effect. Is the
>> procedure I outlined reasonable, or is there another way?
>>
>
>I'm willing to be corrected on the setting issue, but observation, and
>consideration of the problem, suggest that it is every bit as "rule of thumb"
>as any other plumbing job, and about two thirds shut would be a reasonable
>guess since you want to mimic the load of the system in very rough terms.
>I've certainly never seen a plumber do anything than give the bypass gate a
>good twirl!
Can't agree with that as a general rule. It might be OK on a compact
installation with lots of pump head in hand, but in my rambling house, I
need every inch of head and the last radiator is still not getting
enough. Before you ask, no I didn't instal it and it does need
improving. It's on the list!
>> Particularly on procedures for setting the bypass, and to debate the
>> case of not having a bypass at all.
>
>Well, since the maximum downside of having an open bypass is a slight loss of
>head, which in a suitably "designed" system will not impact on system
>performance, I don't think that once installed there is any point in either
>removing or closing it. Experimenting with our installation indicates that
>there are no evident noise effects of having it open or closed under normal
>running, nor any perceptible improvements in performance with it closed.
Yes, if you already have one it's worth setting it up, and the best
position might well be fully closed. My question is, can it be entirely
omitted on new installations when a well protected boiler is installed.
Do we now agree that protecting the pump is not an issue.
Phil Addison
Phil Addison
<dumb...@my-deja.com> wrote:
>In article <3A77DD80...@hursley.ibm.com>,
>I would suggest you don't need this bypass because as you are assuming
>with a three way valve there is always a route for the flow - assuming
>that there is at least one radiator that doesn't have a TRV fitted.
The rule is that you need such a not TRV'd radiator if you *don't* have
a bypass.
>If the the bypass is open then a large proportion of the pumped flow
>will simply return back to the bolier without going through the
>radiators or HW tank; not very efficient.
Quite so. Inefficient from a POV of loosing pump head so distant rads
don't get warm.
>In principle dead heading the pump doesn't necessarily mean that it
>will be damaged.
I dead head my roses!!
>It will simply produce its maximum head at no flow.
>Conversly providing little or no resitance will allow it to run to the
>end of its curve with maximum flow and very little head - how far along
>the curve being dictated by the maximum power of the motor. The only
>provision to this is that when dead headed for a long time the pump
>motor could overheat since there would be little or no flow around the
>motor rotor.
> I would guess that the motor relies on some internal
>circulation for lubricating the bearings and cooling, but then only the
>pump manufacturer would know this. Dead heading for a short time, a few
>minutes or so, shouldn't matter whatever the requirements.
The wet bearing will still be wet, so this won't be a problem.
>In the end though the simple answer should be to close the bypass and
>make sure that there is at least one radiator without a TRV. Or fit a
>pressure dependent bypass valve as suggested elsewhere.
Possibly, but it's not the answer to the topic question, which is "how
do you set up a by-pass valve?"
Phil Addison
Jim Easterbrook
> On Thu, 1 Feb 2001 20:51:14 +0000, in uk.d-i-y j...@easter.mersinet.co.uk
> (Jim Easterbrook) wrote:
>
> >4/ close all TRVs and manual rad valves, ensure boiler temperature
> >difference is within limits
>
> Huh? With all valves shut the boiler will cycle on its overheat stat.
> The return will be non-existent (i.e. cold).
If your boiler is happy with a non-existent return, i.e. no flow, then I
don't think you need a bypass. My boiler's instructions make it very
clear that an adequate flow rate must be maintained whenever it is on
(i.e. powered up and able to fire). With all TRVs shut this means that a
small volume of water circulates through the boiler - pump - bypass -
boiler circuit. When this water cools sufficiently, the boiler fires and
reheats it. The temperature difference during this re-heating is what I
measure as the worst case temperature difference, and ensure is below
the manufacturer's upper limit.
> There is no way of estimating the flow numerically. All you can say is
> that there is some, and possibly that it is high enough if the boiler
> stat does not cut out.
The heat input from the boiler when firing is known, the heat capacity
of water is known, therefore the temperature rise across the boiler
(when firing) allows you to derive the flow rate through it. Surely?
> >Although with all the valves closed, you don't get much time to
> >measure this before the boiler stat shuts it down.
>
> No indeed. And no flow to 'estimate' either!!
The flow is through the bypass. That's what it's there for.
> >Fortunately I bought
> >two multimeters with thermocouples.
>
> For balancing I imagine, not for flow-rate.
Both. My boiler's guarantee is invalidated if they find signs of
kettling or overheating. A properly set up bypass is a means of avoiding
this. If the temperature difference across the boiler is low enough, the
flow rate is high enough.
Phil Addison
(Jim Easterbrook) wrote:
>Phil Addison <phi...@bigfoot.com> wrote:
>> On Thu, 1 Feb 2001 20:51:14 +0000, in uk.d-i-y j...@easter.mersinet.co.uk
>> (Jim Easterbrook) wrote:
>> >4/ close all TRVs and manual rad valves, ensure boiler temperature
>> >difference is within limits
>> Huh? With all valves shut the boiler will cycle on its overheat stat.
>> The return will be non-existent (i.e. cold).
>My boiler's instructions make it very
>clear that an adequate flow rate must be maintained whenever it is on
>(i.e. powered up and able to fire).
Such an instruction, and it seems the norm, does put the onus on the
installer to fit and adjust a means of bypass. I have been wondering if
the more sophisticated boilers could protect themselves against no-flow
conditions by simple temperature measurements, and then hopefully state
in the instructions that they don't have a minimum flow requirement.
>With all TRVs shut this means that a
>small volume of water circulates through the boiler - pump - bypass -
>boiler circuit. When this water cools sufficiently, the boiler fires and
>reheats it. The temperature difference during this re-heating is what I
>measure as the worst case temperature difference, and ensure is below
>the manufacturer's upper limit.
>
>> There is no way of estimating the flow numerically. All you can say is
>> that there is some, and possibly that it is high enough if the boiler
>> stat does not cut out.
>The heat input from the boiler when firing is known, the heat capacity
>of water is known, therefore the temperature rise across the boiler
>(when firing) allows you to derive the flow rate through it. Surely?
Temp Rise = Heat input x Sp. Heat x Mass Flow Rate
and the Mass Flow Rate is???
>> >Although with all the valves closed, you don't get much time to
>> >measure this before the boiler stat shuts it down.
>>
>> No indeed. And no flow to 'estimate' either!!
>
>The flow is through the bypass. That's what it's there for.
I guess your description was just too succinct for me. You did say
"3/ back off bypass till it closes"
which is why I said "Huh??" when in 4/ you closed the rads as well.
>Fortunately I bought two multimeters with thermocouples.
> If the temperature difference across the boiler is low enough, the
>flow rate is high enough.
I see what you're getting at now. You are actually measuring the
differential temperature as the overall temperature rises. I agree with
the above line as you have it, but I'm not sure what 'low enough' would
be. A typical boiler is specified to raise the water by 10C at its
specified *full* flow rate. The flow through a *manual* bypass should be
much less than this with a consequent much higher temp rise. I don't
know about an automatic bypass - I suppose if it fully opens the flow
may approach the full rate. So I agree your method seems a valid one for
your system.
Sorry for the misunderstanding - I was really focussed on a manual
bypass which is what my original question was about.
--
Phil Addison
UK DIY FAQ is at http://pages.eidosnet.co.uk/~ukdiy/
Short-cut via http://www.uk-diy.org/
Mark Dumbrill
> On Thu, 01 Feb 2001 19:47:19 GMT, in uk.d-i-y Mark Dumbrill
> <dumb...@my-deja.com> wrote:
> >In article <3A77DD80...@hursley.ibm.com>,
>
> >I would suggest you don't need this bypass because as you are assuming
> >with a three way valve there is always a route for the flow - assuming
> >that there is at least one radiator that doesn't have a TRV fitted.
>
> The rule is that you need such a not TRV'd radiator if you *don't* have
> a bypass.
>
radiator is not turned off. Even in a house with no TRVs I guess all the
radiators could get turned off. Then with a three way valve on CH only there
is no flow around the boiler. I must admit I hadn't considered the case of
all the radiators being turned off before it was mentioned here. If this
were to happen then the problem is going to be in the boiler, kettling,
overheating, etc. What I don't believe you have to worry about is the pump
unless it was left running for a long time.
>
> >It will simply produce its maximum head at no flow.
> >Conversly providing little or no resitance will allow it to run to the
> >end of its curve with maximum flow and very little head - how far along
> >the curve being dictated by the maximum power of the motor. The only
> >provision to this is that when dead headed for a long time the pump
> >motor could overheat since there would be little or no flow around the
> >motor rotor.
>
> > I would guess that the motor relies on some internal
> >circulation for lubricating the bearings and cooling, but then only the
> >pump manufacturer would know this. Dead heading for a short time, a few
> >minutes or so, shouldn't matter whatever the requirements.
>
> The wet bearing will still be wet, so this won't be a problem.
>
the bearings and motor might not work effectively. They'd obviously still be
wet, but with poor circulation the bearings/motor could overheat. With a
motor that's only rated for less than a 100W this might not be a big
problem - but in the end that's the most likely mechanism for damaging a
pump of this type.
Mark Dumbrill
Phil Addison
<ma...@dumbrill.com> wrote:
>Phil Addison <phi...@bigfoot.com> wrote in message
>news:U7p6OkErm0dgyL...@4ax.com...
>>
>> a bypass.
>>
>As has been pointed out earlier you can't guarantee that the non TRV'd
>radiator is not turned off. Even in a house with no TRVs I guess all the
>radiators could get turned off.
Ok, the rule is that you need such a not TRV'd radiator if you *don't*
have a bypass and you shut it off at your own risk. This is only meant
to a word of warning to folk who might willy nilly retro-fit TRVs.
Ideally they should be advised to make sure they have a (automatic?)
by-pass if they are doing this. As for initial fit... how can you get
the trade to do it properly? Is there a Standard - I bet there is not.
>If this
>were to happen then the problem is going to be in the boiler, kettling,
>overheating, etc. What I don't believe you have to worry about is the pump
>unless it was left running for a long time.
I know of one case where all the rads are TRV'd and when they close down
a terrific water hammer starts up and the family rush to open up one of
the TRV's and turn down the hall stat.
>> > I would guess that the motor relies on some internal
>> >circulation for lubricating the bearings and cooling, but then only the
>> >pump manufacturer would know this. Dead heading for a short time, a few
>> >minutes or so, shouldn't matter whatever the requirements.
>>
>> The wet bearing will still be wet, so this won't be a problem.
>>
>Not necessarily because with no forward flow the internal circulation round
>the bearings and motor might not work effectively. They'd obviously still be
>wet, but with poor circulation the bearings/motor could overheat. With a
>motor that's only rated for less than a 100W this might not be a big
>problem - but in the end that's the most likely mechanism for damaging a
>pump of this type.
Manufacturers should state if there are such operating limits on their
pumps. I would assume if they don't warn against it, it's OK.
Jim Easterbrook
> On Fri, 02 Feb 2001 18:31:23 GMT, in uk.d-i-y j...@easter.mersinet.co.uk
> (Jim Easterbrook) wrote:
>
> >My boiler's instructions make it very
> >clear that an adequate flow rate must be maintained whenever it is on
> >(i.e. powered up and able to fire).
>
> Such an instruction, and it seems the norm, does put the onus on the
> installer to fit and adjust a means of bypass. I have been wondering if
> the more sophisticated boilers could protect themselves against no-flow
> conditions by simple temperature measurements, and then hopefully state
> in the instructions that they don't have a minimum flow requirement.
I think the problem arises with low water content boilers. If the flow
stops, the temperature rise in the heat exchanger might be a bit too
rapid for a thermostat (on the outside) to respond in time. There are
too many variables (system volume, pump capacity, radiator circuit
resistance and so on) outside the boiler manufacturer's control to avoid
putting the onus on the system designer / installer, in my view.
> I guess your description was just too succinct for me. You did say
> "3/ back off bypass till it closes"
> which is why I said "Huh??" when in 4/ you closed the rads as well.
Sorry, I should have made it clear I was talking about an auto bypass.
This wants to be adjusted to be just closed when the rad valves are
open, but to start opening itself as rad valves are closed. When all rad
valves are closed it will be fully open, keeping up the flow rate
through the boiler. It does seem to do this rather well, and is not very
expensive.
Tim Hardisty
wrote:
>On Thu, 01 Feb 2001 09:37:54 GMT, in uk.d-i-y t...@hardisty.co.uk (Tim)
>wrote:
>>FWIW, I fitted an automatic differential bypass valve. ..
>>It is normally closed, but opens up when the pressure difference
>>exceeds a preset limit, between 0.1 and 0.6 bar IIRC.
>
>I'm not clear what this means. Is it adjustable to open somewhere in
>that range, and then opens within .05 bar (from below) of the setting?
Yes and no - and/or I think so! This could turn into a semantic
argument (which, BTW, is a daft game IMHO).
It is adjustable with a calibrated dial to open at a specific pressure
differential - if set to "2" it opens at 0.2 bar (calibration error,
tolerances etc not withstanding). The point it opens _is_ the set
pressure.
The actual pressure it is set-up to open at will be 0.05 bar higher
than that determined during set-up - this difference arising from the
half a turn the dial is backed off.
So, if set to "2" it opens at 0.2bar pressure difference. The set-up
may have determined 0.15 as the point at which the valve starts to
open - so back off half a turn to set it to 0.2 bar.
Difficult to explain that! If needs be, I'll try again if I've still
not made my point clear :)
Regards,
Tim Hardisty.
Phil Addison
Hardisty) wrote:
>Difficult to explain that! If needs be, I'll try again if I've still
>not made my point clear :)
No need - I've got it now, thanks.