Dear Rakesh,
Mr. Macharla has thrown some light on the fluid dynamics part of it. The basic understanding is whenever there is a change of direction turbulence or mixing is created in fluid. As a result the measuring device receives fluctuations in flow and unable to sense the actual flow velocity. So we need some straight lengths both upstream and downstream of the instrument. The flow pattern depends on the basic pipe routing including type of fitting or valve. If there are two bends preceeding a flow measurement device the distance between these bends also affect the flow pattern. It also depends on the sensitivity of instrument otherwise known as Beta ratio. It is a normal practice to select 0.7 as Beta ratio for selecting upstream and downstream distance before getting actual values from vendor. Later on when actual Beta ratios are obtained the upstream and downstream distances can be checked. You may please look into ISO 5167 which is the relevant standard followed.
Regards
B. Gopakumar
Telefon: 91-6347990 Ext 1543
Telefax: 91-6347512
mailto:B.Gopakumar@pg.siemens.co.in
-----Original Message-----
From: V.R.H.P. Macharla [mailto:macharla777@yahoo.com]
Sent: 26 November 2001 10:21
To: PipingDesign@yahoogroups.com
Subject: Re: [PipingDesign] Flow Elements - Recommendation
Rakesh,
I suppose you are looking for the philosophy behind
deciding the straight lenths before and after flow
element.
It links to theory of Boundary Layer and Secondary
Flows in fluid path. After any bend, there will be
secondary flows induced near the pipe wall...it
proportionately effects local velocity/pressure of the
fluid. Considering the length of pipe required to
reach to full velocity of liquid, standard straight
lengths ahead of Flow Element are established.
For after FE, also staright length is required to
reach to full velocity after the Vena Contracta
condition at FE.
Hope it clarifies upto some extent.
Macharla
--- saud <saud@bushnak.com> wrote:
> Dear Rakesh
>
> The standard recommended dimensions upstream and
> downstream of a flow
> measuring instrument sensor (paddle wheel type etc.)
> is 10 times the
> diameter of the pipe (D) for upstream and 5 D for
> down stream.
>
> With latest magnetic flow meters, this dimension is
> reduced to 5 D and 2 D
> respectively.
>
> Regards
>
> Saud
> ----- Original Message -----
> From: "RAKESH PATIL" <rpa@epc.ltindia.com>
> To: <PipingDesign@yahoogroups.com>
> Sent: Friday, November 23, 2001 6:15 PM
> Subject: [PipingDesign] Flow Elements
>
>
> > ** High Priority **
> >
> > Can anybody give me the basic criteria behind
> deciding the upstream and
> downstream dimensions between Flow Element and
> connecting Fitting /
> accessories in piping?
> >
> > regards,
> > rakesh
> >
> >
> >
> >
> ___________________________________________________
> > If you want only site update notifications, go
> here:
> > mailto:PipingDesignNews-subscribe@yahoogroups.com
> > _____________________
> > Piping Design Central
> > [ http://www.PipingDesign.com ]
> > ________________________________________
> > We hope you find this newsletter useful.
> > If not, you can unsubscribe at:
> > PipingDesign-unsubscribe@yahoogroups.com
> > Post message: PipingDesign@yahoogroups.com
> > Subscribe: PipingDesign-subscribe@yahoogroups.com
> >
> > Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
> >
> >
>
> Disclaimer :
> Our server is equipped with a Virus protection
> system and scans most recent viruses automatically
> prior to receiving / sending messages. We have taken
> maximum possible care by installing state-of-the-art
> protection programs to safeguard against viruses.
> However, since there can be no absolute protection,
> we are not responsible for any damage, loss of data
> and/or hardware damage caused by viruses to the
> recipients of mails sent or received through our
> server.
>
> Bushnak Water Group
> www.bushnak.com
>
>
>
>
>
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/ Received on Mon Nov 26 01:42:00 2001
This archive was generated by hypermail 2.1.8 : Mon Oct 27 2008 - 20:23:55 EDT