Steve:
I've worked with a wide variety of flowmeters in a wide variety of services, including the mag type, in smaller pipes, over the years. Based on what I've seen, you are unlikely to find any ISO-type standards or reports on the net effect of upstream and downstream distances, as the standard distances are based on most conservative, reliable performance.
The issues are that any effects are going to depend on a wide variety of parameters, all of which will effect turbulence and pressure drop: pipe roughness, fluid viscosity (and things that effect viscosity, such as temperature), means of connection on upstream/downstream fittings, type of fitting/disturbance, etc. With low viscosity liquids like water, good quality butt-welded fittings, reasonable turning radii, and so forth, for the conditions you describe (half the recommended distances), you should be consistent and reasonably close (within 5-7%) of the reading that you would get with the recommended distances.
... Bruce D. Bullough ...
Sebesta Blomberg & Associates, Inc.
2381 Rosegate
Roseville, MN 55113 USA
+ 651-634-7344
www.sebesta.com
-----Original Message-----
From: Steve McKenzie [mailto:mechproj@xtra.co.nz]
Sent: Thursday, February 17, 2005 4:33 AM
To: <a href="/group/PipingDesign/post?postID=4xP36nV_jOxp9oTUaanGdgvJDwokPBZzXLVoH6c6a1CutI_Ufj15Z24LwPVW0PCOTVWOvyj0gbafRT3Y3BeMbSbkHYHE">PipingDesign@yahoogroups.com</a>
Subject: RE: [PipingDesign] querry.... [bcc][faked-from][mx][spf][bayes]
Hi Al
thanks for your reply.
If you have a look through ISO5167, it provides expected accuracy limits
for a number of configurations; presumably derived from physical tests.
How? I don't know but would guess comparative measurement using two
flowmeters; one in an ideal location and one in a test location but both
in the same stream at the same time.
What I am after is a similar set of measurements where the upstream and
downstream distances are varied and the accuracy measured, instead of
the accuracy being nominated and the distance stated. From a set of test
results showing the scatter , a simple statistical evaluation will give
an reasonable idea of the level of confidence in the reading of any
specific geometrical configuration. The trick will be deciding on the
shape of the confidence curve.
I see no difficulty in this which is slightly different from your case
where it appears you were after predictability in flow symmetry
immediately after a flow disturbance. In my case I will live with the
lack of predictability; I just want to quantify it. By the way I failed
statistics at university; partly through boredom, partly through
thickness and the balance being that the lectures clashed with my
snooker school. In other words, I went to a few lectures, couldn't
understand them or stay awake so I went and played snooker instead.
In simplistic terms, if a flowmeter manufacturer can assure a given
level of accuracy with one set of up/down diameters, then surely another
level of accuracy must be possible with a different up/down diameter
set. +/- 50%? I don't care; I just want to know what it is. Two standard
deviation confidence will do just fine.
Cheers
Steve
-----Original Message-----
From: aluser2 [mailto:alwynk@shaw.ca]
Sent: Thursday, February 17, 2005 10:50 PM
To: <a href="/group/PipingDesign/post?postID=4xP36nV_jOxp9oTUaanGdgvJDwokPBZzXLVoH6c6a1CutI_Ufj15Z24LwPVW0PCOTVWOvyj0gbafRT3Y3BeMbSbkHYHE">PipingDesign@yahoogroups.com</a>
Subject: Re: [PipingDesign] querry....
steve
it is possible to look at the outcome of this in
a CFD program and we have had good success eliminating bend flow
perturbation (upsetting pneumatically driven control valves) with the
tools. With far less than 5D after a complex bend Geometry alterations
were still able to get stable control at the control valve and still was
predicted by the analysis. There was considerable resistance in both
instances to any change , but this was unreasonable and they were
desperate. in the one case they switched from pneumatic to electronic
control, although this alone was insuficient as the shaking and
cavitation was extreme.
in the other case with similar tight and out of planeY geomtery almost immediately followed by a 90 and then 3D to the valve the control valve settled right down after flow smoothing by geometry alteration
i dont think you will find any data as each case is far too custom. What you are looking for is the flow behaviour and that is geometry dependent.
>
> Hi Robin
>
> from your burst of excellent and detailed responses to queries, it
> appears you have some time on your hands.
>
> My question#1 is when does your boss get back from his vacation?
>
> Question #2, I have to bang a magflow into an existing line and intend
> using the 2/3 upstream, 1/3 downstream that I normally use. Trouble is
> that I wont have my normal minimum 10D up and 5D down probably more
> like 5D up and 2.5D down, closer to your numbers. The disturbances at
> each end are 1.5D bends, same plane. Line size is probably 150NB and
> flow rate around 170m^3/h, water. The metering is for process control,
> not accounting, so accuracy isnt hellish important but it would be
> comforting to know the expected accuracy. Have you or anyone else seen
> any data on the effect on accuracy when departing from "the rule"? I
> expect a slightly higher reading than actual at higher flow rates
> (due to spin), but someone has surely done a bit of research on this.
> Am up to my ears in rules of thumb and would prefer a bit of
> supporting research data. Have perused ISO5167, but it doesn't really
> apply to an induction driven meter.
>
> Personally, I feel this question may take a couple of hours of digging
> to find the answer, which I am sure would be of use to a number of
> list members. I suggest that any member who makes a "crime" against
> this group be "fined" by being required to provide a well-researched
> and referenced (1 page minimum) response to a nominated question, such
> as this (question #2, or course), within one week. It would improve
> the knowledge base of this group, and reset the clock. Paul could
> judge if such a crime has occurred. He would like it as he likes
> throwing his weight around. Whaddayathink?
>
> Cheers
>
> Steve
>
>
>
>
>
>
>
> -----Original Message-----
> From: Robin Badcock [mailto:rbadcock@vision.net.au]
> Sent: Thursday, February 17, 2005 1:57 PM
> To: <a href="/group/PipingDesign/post?postID=4xP36nV_jOxp9oTUaanGdgvJDwokPBZzXLVoH6c6a1CutI_Ufj15Z24LwPVW0PCOTVWOvyj0gbafRT3Y3BeMbSbkHYHE">PipingDesign@yahoogroups.com</a>
> Subject: Re: [PipingDesign] querry....
>
>
>
> Vaibhav,
>
> There exists numerous distances for devices that are inserted within
> the pipeline and these vary somewhat to the duty they are to perform,
> i.e. paddle switches may have a requirement of 5 times pipe diameter
> of straight pipe up-stream and down-stream from the point of
> installation. Sensitive propeller type instruments may call for 10
> times pipe diameter. Generally speaking the manufacturer will provide
> you with guidelines on the correct installation. These guidelines are
> in the interest of having the device operate, and achieve the base
> line accuracy that the maker claims on their literature.
>
> This does not answer your question fully. The reason for the straight
> pipe sections is for the reduction of possible turbulent flow that may
> affect the devices operation.
>
> In answer to you query on the straight pipe sections associated with
> check valves, we have witnessed first hand non return valve mechanical
> failure due to excessive turbulence when installed too close to the
> down stream side of a pressure sustaining valve. The end result was
> total internal valve failure within two years. This less than ideal
> installation was due to space constraints. One solution was to install
> a more robust valve and be prepared to live with the extra
> maintenance. The correct solution would have been to observe a greater
> distance of straight pipe work, down stream of the sustaining valve,
> if space would have permitted.
>
> If you goggle pipeline turbulence you may turn up some useful info.
> Perhaps Paul has a link or some information on the Piping Design web
> page.
>
> The orifice plate that you may be manufacturing yourself for head
> loss/flow regulation will also be subject to straight piping
> requirements for the same reason of attaining reliable and repeatable
> regulation. If installed into a turbulent pipeline area variations may
> well exist. Software is available from the likes of Dorot, Bermad and
> quite likely other control valve manufactures and although I have not
> checked, software and information is likely to be freely available on
> this subject. This was certainly the case some years ago.
>
> I hope that this helps.
>
> Cheers for now.
>
> Robin Badcock
>
> ----- Original Message -----
> From: <VAIBHAV_DEVDHAR@ril.com>
> To: <PipingDesign@yahoogroups.com>
> Sent: Wednesday, February 16, 2005 11:22 PM
> Subject: [PipingDesign] querry....
>
>
> >
> >
> >
> >
> >
> >
> >
> >
> > Can any one please tell the required straight upstream and
> > downstream requirement for Restriction Orifice? Please note that
> > this orifice is only for pressure drop purpose and not for flow
> > measurement. What could be the reason for this requirement?
> >
> > Has anyone come across the straight length requirement for check
> > valves, especially swing check valves? Why is this needed?
> >
> > Request members to throw some light on this.
> >
> > Regards
> >
> > Vaibhav Devdhar
> >
> > [Non-text portions of this message have been removed]
> >
> >
> >
> >
> >
> >
> >
> > =========================================
> > PipingOffice - Excel Spreadsheets for Piping Calculations
> > <a href="http://www.pipingoffice.us/">http://www.pipingoffice.us/</a>
> > =========================================
> > Main site: <a href="http://www.pipingdesign.com">http://www.pipingdesign.com</a>
> >
> > Yahoo! Groups Links
> >
> >
> >
> >
> >
> >
> >
> >
> >
>
>
>
>
>
> =========================================
> PipingOffice - Excel Spreadsheets for Piping Calculations
> http://www.pipingoffice.us/ =========================================
> Main site: <a href="http://www.pipingdesign.com">http://www.pipingdesign.com</a>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
>
>
>
> =========================================
> PipingOffice - Excel Spreadsheets for Piping Calculations
> http://www.pipingoffice.us/ =========================================
> Main site: <a href="http://www.pipingdesign.com">http://www.pipingdesign.com</a>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
<a href="http://www.pipingoffice.us/">http://www.pipingoffice.us/</a> =========================================Main site: <a href="http://www.pipingdesign.com">http://www.pipingdesign.com</a>
Yahoo! Groups Links
Yahoo! Groups Links Received on Thu Feb 17 11:35:00 2005
This archive was generated by hypermail 2.1.8 : Tue Mar 04 2008 - 11:40:39 EST