Hi Ken
Unless you have completely independent power supplies, it would be essential to consider the case when all pumping stations lose their power supply simultaneously.
Am staggered that a district CH-W distribution system can be made to pay its way; would love to see the design occupancy models, and the energy costs. I assume its somewhere that gets very hot.
The idea of using a 1050NB manual close valve on this system is a little disturbing. You may want to consider the closing time.
However I find it difficult to believe you could create a "system" surge event of any significance, only pump initiated surges. If this is the case, then assuming the "plant"is the refrigeration and pumping plant,your suggestion of having the plant designers control the surge effects appears practicable. You will need to negotiate and agree with them the operating conditions unless conditions are stipulated by a regulating authority.
If the air handlers are at the top of tall buildings and static lifts are a significant component of the pump head (i.e. no pressure break heat exchangers) , then Geoffs high pressure spike will require careful consideration. But if the system resistance is predominantly frictional, then some local attention to the pumps may be all that is required.
I have not done a district scheme, but have done a few plant-wide schemes with third party users. In these cases, I have told they users what they will get in terms of supply temperature, normal pressure, available differential pressure, maximum/min permitted excursion pressure, max permitted return temp, and the users tell me what their demand is. It seemed to work OK, but considerable judgement was necessary as the users had a very poor preliminary idea of their requirements.
It is difficult to comment on the wall thickness of the pipe without knowing the operating conditions.
Cheers
Steve
-----Original Message-----
From: PipingDesign@yahoogroups.com [mailto:PipingDesign@yahoogroups.com]
On Behalf Of ken eppleston
Sent: Monday, January 16, 2006 4:27 PM
To: PipingDesign@yahoogroups.com
Subject: RE: [PipingDesign] Surge and Water Hammer
Thanks for the info,
This would serve air handling systems only and any refrigeration required for a retail area would be served from a local plant- most likely air cooled. But you are right if a building went offline there would be a few unhappy building tenants.
The network is designed in such a fashion as to provide redundancy and reliability in the event of a plant failure so unless there is a catastrophic failure (or a local failure) cooling capacity is always available.
I guess from our perspective we are designing the network and not the plants so we need to ascertain what responsibility we should take of another engineers plant design and the way it is operated. Should it not be the plant designers liability that if he operates the network in such a fashion as to induce pipe surge or water hammer?
Not to say we should stick our heads in the sand and ignore it, but manage it so that the plant designer understands it is their responsibility to conduct the analysis.
The manually operated valves are DN1050 with gear driven spindle, so not a lever type valve and cannot be shut off quickly.
The nature of the diversity on the site is such that the network will probably never be at peak demand running flat out at design pressure. The control of the plants will dictate how the system reacts under a pumping plant failure and there are actually 4 plants pumping into the same network loop so if 1 was to fail how will a major pressure spike or negative pressure be induced? The remaining 3 pumping stations will ramp up (slowly) to make up the short fall in system pressure caused by the plant failure.
At each take off there are regulating valves with a minimum set point to avoid the issues of stagnation in dead legs this implies that there are no fast acting shut off valves at each take off.
Also, using schedule 40 steel as opposed to standard pipe obviously makes the supply more expensive but are there other issues for example with the supply of valves and flanges?
Any thoughts??
Geoff Stone DD&D Australia <blenrayaust@yahoo.co.uk> wrote:
Steve,
You may well be right but until you look at the system how do you know?
If the system is working at full flow and there is power trip there will likely be a negative pressure wave followed by an uncontrolled high pressure spike. The negative pressure wave could be sufficient to collapse the pipe. The resultant high pressure will most likely contained in the pipe. The physical properties of carbon steel under high strain rates being improved (Piping Design-Kellogg).
It is a disrict coling scheme with many users. However a pressure spike would cause noise to users. If a sub header was to fail anbd a building went off line the owner would not be a happy Vegemite. The risks have to be eveluated. Likelihood and consequences. If the consequence is loss of cooling water to a shopping centre, loss of trade and foodstuffs spoiling you are in dire straits.
The point I make is that without using the tools available to evaluate something your "knowledge is meagre and unsatisfactory" Lord Kelvin.
Geoff
Steve McKenzie <Mechproj@xtra.co.nz> wrote:
Hi Robin
Have you changed lifestyle?
I am about to need some help on dust suppression.
Have you done anything in this regard?
This is for Kens digestion and Geoffs indigestion.
Geoff will cane me for this, but for water supply I seldom worry about
surge protection at concept phase other than to throw 15% on the design
pressure and 10 - 20% on pumping station costings. The iterative nature
of design normally requires an estimate to be made on secondart issues.
Surge is a secondary effect and can only be evaluated after the primary
system is laid out. Surge anticipators are good on a force mains, but
most (true) networks have so much frictional diffusion and capacitance
that they never reflect anything much. This aint the case with force
mains which can blow your head off. With a truly closed system, and I
have never done a closed network (assume its district heating/cooling),
things will change a little but diffusion and unequal conduit lengths
should tend to stop things getting out of hand. Also with a network, the
serious events are supply (pump) generated and the likelihood of a
widespread (valves closing
simultaneously) flow rejection is normally impossible. Experience with
previous open circuit projects suggests to me that you may need some air
vessels at each pump station at preliminary costing stage, and these
will be dropped off as the surge model matures. If the system is a
district heating or cooling system then you will need some volume
compensation capability anyway so the surge vessels may double as
pressure accumulators for volume change; thus reducing cost. My gut feel
is that if there are heaps of small demand users, no major elevation
changes and no long force mains (pressure pipes with no active
branches) then eventual provisions for water hammer will be bugger all,
and water hammer will be relegated to a secondary issue.
It pays to remember that the majority of major liquid distribution networks were commissioned prior to the proliferation of water hammer software, and most operate acceptably with regard to transients.
Cheers
Steve
-----Original Message-----
From: PipingDesign@yahoogroups.com [mailto:PipingDesign@yahoogroups.com]
On Behalf Of Robin Badcock
Sent: Friday, January 13, 2006 11:03 PM
To: PipingDesign@yahoogroups.com
Subject: Re: [PipingDesign] Surge and Water Hammer
Ken,
As an added safety feature of the piping network will there be surge anticipation valves fitted.
Cheers,
Robin Badcock
Irrigation Design Consultant
Badcock Irrigation Services
4154 Meander Valley Road
Deloraine, Tas. Australia. 7304
Ph. 03 63623003
Fax. 03 63622977
Sent: Friday, January 13, 2006 10:02 AM
Subject: [PipingDesign] Surge and Water Hammer
> This mostly directed at Geoff but all advice is welcomed,
>
> We have a large closed network utilising insulated steel pipe and I am
> trying to understand the likelihood or risk associated with a pressure
> surge.
>
> This network is entirely buried and we have not been commissioned to
> undertake the pumping plant design- so we have little control of valve
> selection and/or control within the plants.
>
> There will be 3 pumping stations each pump is variable flow and all
> valves in the network are manually operated large diameter butterfly
> valves. Roughly there will be 20 pumps in each pumping station.
>
> My limited understanding of surge and water hammer is that it relies
> on a fast change in pressure (by valve closure or loss of pumps etc)
> in the network which propogates through out the system causing if not
> catastrophic failure then fatigue.
>
> The overall system pressure will modulate in accordance with demand
> but this will be controlled by VSD's on the pumps, also given that the
> entire network is completely restrained by the soil and that there is
> no real risk that a valve can be shut quickly...Is surge and water
> hammer still a potential risk on the system and why?
>
> The size of the network dictates that any ramping up or down of pumps
> will not induce a rapid change in the network pressure. And even if
> one pumping station goes down because of power failure the other two
> will be designed to pick up the slack. My thoughts are that this still
> may induce a rapid change in pressure??
>
> Can managing the operation (including the event of a failure) of the
> network in such a way as to avoid any rapid pressure variations be
> used to eliminate the risk or surge and water hammer?
>
> Any ideas or comments??
>
>
>
>
>
>
>
>
> =========================================
> PipingOffice - Excel Spreadsheets for Piping Calculations
> http://www.pipingoffice.us/ =========================================
> Main site: http://www.pipingdesign.com
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
>
http://www.pipingoffice.us/ =========================================Main site: http://www.pipingdesign.com
Yahoo! Groups Links
-- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.371 / Virus Database: 267.14.17/227 - Release Date: 1/11/2006 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.371 / Virus Database: 267.14.17/227 - Release Date: 1/11/2006 ========================================= PipingOffice - Excel Spreadsheets for Piping Calculations http://www.pipingoffice.us/ ========================================= Main site: http://www.pipingdesign.com Yahoo! Groups Links Design Detail and Development (a division of Blenray Pty Ltd) Mail Address PO Box 1351 Castle Hill NSW 1765 Australia Tel Mob 0402 35 2313 Office 02 8850 2313 AH 02 8850 2324 We specialise in pipe network and waterhammer analysis, pipe stress analysis, the design of buried pipelines and thermoplastic pipe systems. [Non-text portions of this message have been removed] ========================================= PipingOffice - Excel Spreadsheets for Piping Calculations http://www.pipingoffice.us/ ========================================= Main site: http://www.pipingdesign.com SPONSORED LINKS Mechanical engineering Mechanical engineering jobs Cad drafter Process piping Mechanical --------------------------------- YAHOO! GROUPS LINKS Visit your group "PipingDesign" on the web. To unsubscribe from this group, send an email to: PipingDesign-unsubscribe@yahoogroups.com Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. --------------------------------- --------------------------------- Do you Yahoo!? Yahoo! News: Get the latest news via video today! [Non-text portions of this message have been removed] ========================================= PipingOffice - Excel Spreadsheets for Piping Calculations http://www.pipingoffice.us/ ========================================= Main site: http://www.pipingdesign.com Yahoo! Groups Links -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.1.371 / Virus Database: 267.14.18/230 - Release Date: 1/14/2006 -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.371 / Virus Database: 267.14.18/230 - Release Date: 1/14/2006Received on Mon Jan 16 05:38:00 2006
This archive was generated by hypermail 2.1.8 : Mon Oct 27 2008 - 20:24:10 EDT