"Malcolm Reeves" wrote in message
...
On Thu, 23 Sep 2004 09:33:06 +0100, Andy Hall
wrote:

On Thu, 23 Sep 2004 09:18:51 +0100, Malcolm Reeves
wrote:



Between a certain range of flow/pressure, the Alpha behaves
conventionally. When the flow falls below a certain point,
that is
detected and the power is reduced to reduce the head.

Not according to the alpha data sheet which says:

Electronically controlled, the pump can, by means of a
selector switch located on the terminal box, be set to:
. 2 constant pressure curves
. 2 proportional pressure curves
. 3 fixed speed curves.

So either constant head (I think I said that), proportional head (head
goes down with lower flow), or 3 normal pump modes.

You need to look at the technical manual for the Alpha.

I am looking at the pdf from the grundfos website
alpha_data_booklet.pdf, 16 pages with graphs etc. from which the quote
above is taken - page 6.

Protection of standard pump nuts can be achieved by a bypass.

Possibly, but there's no point.

You need auto bypass for standard pump - I was being humorous
referring to your comment it would scream its nuts off.

The heat flow from the boiler is the flow, kg/s, and the difference
across the boiler.

Only while the boiler is firing. At the point where the heat used is
less than that produced it will begin to cycle.

Of course but what is the relevance of that. It will always cycle
when Watts in Watts out. We are discussing TRVs and pumps.

I know. The point was that the behaviour is not simple because the
burner is being fired on and off. This has a substantial effect on
the temperatures.

If the boiler is short cycling yes but if it is doing that you need to
fix it since that is very wasteful. If it is on a long cycle then you
can consider it as being in almost steady state. There will be a
steady climb in temperatures but relative temperatures will be similar
during this time.



From the perspective of heat output to the room, it is approximately
proportional to the Mean Water to Air Temperature.

From the perspective of heat delivery to the radiator (which will be
the same since none of it disappears) it will be proportional to the
flow and temperature difference across the radiator.

Agreed (I think I said that).

You can't treat them in isolation though. If the boiler is cycling,
the output temperature is far from being fixed,

Short cycling is a different problem and needs to be fixed. Normal
long cycling is relatively stable temperature difference on an overall
rising base line.

Both effects are driven by
pressure, ones got to happen first then the other won't.

You're missing the point. It is far better to have the pump reduce
output as the heat flow reduces rather than shortcircuiting it back to
the boiler with the pump on full power.

Why? Assuming you have a normal pump set to the right setting. In
fact since there is a recommend minimum speed to stop sludging it
could be argued that a bypass, which keeps the boiler flow constant is
better as it stops the sludge settling in the boiler. Of course if
you correctly treat the water then sludge is not an issue, hopefully.

An alpha would suit something like a coal fired room heater where the
best way to run it is slow continuous release of heat, not
super-burn/off/super-burn/off as you get with normal pump + stat (no
bypass). Also a condensing modulating gas boiler (as long as it could
cope with low flow (coal fires can).

This misses the point completely.

Why? The alpha's modes of constant and proportional head are designed
for system that work best with a variable flow. That has to be
boilers that modulate their output. Also, those boilers that are
happy with a low return temperature, which is another effect of low
flow.

Set it any differently and it's just a fine adjustment standard pump.

Even a fixed boiler has varying output temperature as the burner
cycles. There isn't going to be an issue with the return
temperature being low at low flow rates anyway.

I would tend to agree with you as it would seem reasonable that a low
flow low temperature return is going to quickly warm up in the heat
exchange. However, I'd want to check with the boiler manufacturer
that he was happy with that. I suspect most might prefer the constant
flow arrangement so as to avoid even the only perceived risk of
condensation. So I'd reckon most would recommend normal + bypass
(that's the result I had when talking to HRM). Of course, assuming
this is a standard boiler, not a condensing.

--

Malcolm

Malcolm Reeves BSc CEng MIEE MIRSE, Full Circuit Ltd, Chippenham, UK
, or ).
Design Service for Analogue/Digital H/W & S/W Railway Signalling and
Power
electronics. More details plus freeware, Win95/98 DUN and Pspice tips,
see:

http://www.fullcircuit.com or http://www.fullcircuit.co.uk

NEW - Desktop ToDo/Reminder program (free)



Whoa, guys, you have lost me.!!!


Francis

On Thu, 23 Sep 2004 11:25:13 +0100, Malcolm Reeves
wrote:

On Thu, 23 Sep 2004 08:58:21 +0100, Andy Hall

You can use an automatic bypass purely to take care of the over-run
case.

Are we taking about the same sort of automatic bypass, i.e. pressure
driven or is this an electric bypass? As I fail to see how a pressure
bypass can handle overrun without first handling the run into closed
TRVs first.

Pressure operated. You simply adjust it to the required pressure.

You are assuming that the TRVs are completely closing. This should
not happen if the heat delivery and room thermostat in the non TRV
space are set correctly


If TRVs are open, stat goes off, boiler goes off, pump goes to over
run then water will go to rads (or cylinder).

No. Motorised valves can close and then there is no flow path,
therefore a bypass is useful.


..andy

To email, substitute .nospam with .gl

On Thu, 23 Sep 2004 14:06:38 +0100, "FrancisJK"
wrote:


)



Whoa, guys, you have lost me.!!!


Francis

Your original premise of an Alpha pump and then a bypass should you
need it is a reasonable one....



..andy

To email, substitute .nospam with .gl

"Andy Hall" wrote in message
...
On Thu, 23 Sep 2004 14:06:38 +0100, "FrancisJK"
wrote:


)



Whoa, guys, you have lost me.!!!


Francis

Your original premise of an Alpha pump and then a bypass should you
need it is a reasonable one....



.andy

To email, substitute .nospam with .gl


Don't know about the pump yet. I'll probably go for the alpha, there is an
automatic (differential) by-pass in the circuit already. I can close the
automatic (differential) by-pass fully, if necessary.

OK, Seconds away Round 10!



Francis

"Malcolm Reeves" wrote in message
...
On Thu, 23 Sep 2004 00:21:08 +0100, "G&M"
wrote:


I don't have any direct experience of this, but I was speculating
that -
with an Alpha pump - it may be difficult (or perhaps even impossible?)
to
adjust everything so that the by-pass *only* operates during pump
over-run
conditions.

Oh you can adjust an Alpha to do almost anything. You can even make it
run
like a normal pump if you want though it gets a lot louder then. But I
found that by setting it a 'bit of a turn' more clockwise than is really
required then I can make the by-pass operate.

At last an interesting comment of practical experience. However, I
think the alpha is now operating in almost standard pump mode. The
alpha data sheet curves show for a normal pump change of flow from Q1
DOWN to Q2 gives INCREASE of head from A1 to A2. As expected. But
for alpha Q1 down to Q2 gives DECREASE of head from A2 DOWN to A3.
Which is what you want if don't have an auto bypass.

With an alpha and an auto bypass you have to adjust the alpha to
behave like a normal pump for the bypass to operate, that is flow
down, head up. OK, an alpha has more adjustment than a normal pump
but is the extra cost worth it?

For standard boiler I'd say no as you don't want flow reduction and
low return temperatures with a standard due to the risk of
condensation. For a condensing boiler I would be tempted by an alpha
as with the better adjustment you are more likely to be able to delay
the operation of the bypass and so work sometimes with a lower flow,
lower return temperature, which makes a condenser more efficient.
However I wouldn't like to say for sure that you would recover the
extra GBP35 and alpha costs in the lifetime of the pump.

I've a standard boiler but a very non-standard installtion - boiler is 30
metres from the pump and zone valves in a barn. I don't have many TRVs as
most areas have a prog thermostat and could probably do without the bypass
altogether . But because the system is so large it was suggested here that
come the deep of winter I might need a second pump in the system whereupon I
think the bypass would be a good safety measure.

In most systems the Alpha will save the £35 just in electricity used for
pumping over it's lifetime even if the oil/gas system is no more efficient.

"Malcolm Reeves" wrote in message
...
On Wed, 22 Sep 2004 22:08:16 +0100, Andy Hall
wrote:

On Wed, 22 Sep 2004 21:52:50 +0100, Malcolm Reeves
wrote:

On Wed, 22 Sep 2004 14:00:09 +0100, Andy Hall
wrote:

With an alpha, as TRVs close the flow falls but the head does not
rise
as in normal pumps.

Not exactly.

Between a certain range of flow/pressure, the Alpha behaves
conventionally. When the flow falls below a certain point, that
is
detected and the power is reduced to reduce the head.

Not according to the alpha data sheet which says:

Electronically controlled, the pump can, by means of a
selector switch located on the terminal box, be set to:
. 2 constant pressure curves
. 2 proportional pressure curves
. 3 fixed speed curves.

No. My one has the 3 fixed speed curves and a pot to adjust the
proportional pressure curve almost infinitesimally. There are some other
models introduced since then though.