Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
MEASURING TWO PHASE FLOW 5
- Thread starter ilan7
- Start date
HamishMcTavish
Nuclear
Do you need to measure properties of one phase or both?
No more things should be presumed to exist than are absolutely necessary - William of Occam
No more things should be presumed to exist than are absolutely necessary - William of Occam
There is not any in-line instrumentation capable of accurately measuring two-phase flow of any sort, that I am aware of.
In the process of designing 80 km of steam pipeline, my solution has been to measure the vapor phase at the destination (in this case, 250 injection wells) and subtract it from the total mass into the system. This will only work if you have some point in the system that gives you a baseline of the total flow to the system.
Another method would be to measure the condensate extracted. This is somewhat tricky, in that you'll need to measure the condensate before it's depressurized, since much if it will flash to vapor once depressurized.
Otherwise, you'll need to separate the phases, measure, and recombine the flow.
In the process of designing 80 km of steam pipeline, my solution has been to measure the vapor phase at the destination (in this case, 250 injection wells) and subtract it from the total mass into the system. This will only work if you have some point in the system that gives you a baseline of the total flow to the system.
Another method would be to measure the condensate extracted. This is somewhat tricky, in that you'll need to measure the condensate before it's depressurized, since much if it will flash to vapor once depressurized.
Otherwise, you'll need to separate the phases, measure, and recombine the flow.
-
2
- #5
For geothermal power stations we have a lot of pipelines where measurement of two phase flow is required. No one has come up with a meter but we use a chemical injection method to give spot readings.
We inject two chemicals at a known rate. One chemical dissolves in the water phase and the other becomes a vapor in the steam phase. We them take steam and water samples and the concentrations of the two chemicals with give the steam and water phase flow rates. We get very good results.
We manufacture and sell the equipment if you are interested.
Regards,
Kevin Koorey
Century Resources, Wairakei. PO Box 341 Taupo, New Zealand
Phone +64 7 376 0924, fax +64 7 374 8508
kkoorey@centurydrilling.co.nz
We inject two chemicals at a known rate. One chemical dissolves in the water phase and the other becomes a vapor in the steam phase. We them take steam and water samples and the concentrations of the two chemicals with give the steam and water phase flow rates. We get very good results.
We manufacture and sell the equipment if you are interested.
Regards,
Kevin Koorey
Century Resources, Wairakei. PO Box 341 Taupo, New Zealand
Phone +64 7 376 0924, fax +64 7 374 8508
kkoorey@centurydrilling.co.nz
-
1
- #7
There has been a lot of work done to try to measure two-phase flow in the Oil & Gas industry. At the end of the day even the Ultrasonic Meter (which we all had very high hopes for) was +/- 20-50%. For that kind of spread I can use tea leaves and chicken guts.
The problem is (and will remain) that there is no such thing as steady state multi-phase flow. Every correlation assumes a particular sort of flow (e.g., annuluar, mist, wavy, etc.) and develops empirical data to "measure" that. In the real world flow regime is very dependent on total energy available (at a given instant) and at each band of energy the phases interact differently. I've watched multi-phase flow in clear plastic piping and at the same volume flow rate, pressure, and temperature you'll see every sort of flow regime at a given point in the piping over time. I've despaired of ever getting a measurement methodology that can compensate for very non-steady conditions.
David
The problem is (and will remain) that there is no such thing as steady state multi-phase flow. Every correlation assumes a particular sort of flow (e.g., annuluar, mist, wavy, etc.) and develops empirical data to "measure" that. In the real world flow regime is very dependent on total energy available (at a given instant) and at each band of energy the phases interact differently. I've watched multi-phase flow in clear plastic piping and at the same volume flow rate, pressure, and temperature you'll see every sort of flow regime at a given point in the piping over time. I've despaired of ever getting a measurement methodology that can compensate for very non-steady conditions.
David
zdas04 sounds knowledgeable about this.
We have measured two phase flow indirectly for a steam cooling system where water is sprayed into the steam path to cool it (Turbine bypass).
Normally it is recommended that the mixed flow rate is above saturation. Flow and temperature are measured down stream. If the mixture is below saturation then temperature remains at boiling point no matter what the mixture and flow cannot be effectively measured.
In this case we hade to measure the temperature, pressure and flow for both the coolant and steam. We developed a feed forward algorithm which equates the heat balance using enthalpy. The feed forward control system was linked to the coolant valve position.
Im not sure if this helps your application.
We have measured two phase flow indirectly for a steam cooling system where water is sprayed into the steam path to cool it (Turbine bypass).
Normally it is recommended that the mixed flow rate is above saturation. Flow and temperature are measured down stream. If the mixture is below saturation then temperature remains at boiling point no matter what the mixture and flow cannot be effectively measured.
In this case we hade to measure the temperature, pressure and flow for both the coolant and steam. We developed a feed forward algorithm which equates the heat balance using enthalpy. The feed forward control system was linked to the coolant valve position.
Im not sure if this helps your application.
prewrathrap
Electrical
Without much process data and your description ie 20" line
I have to assume that it is steam line and not super heated.
Coriolis will not work no one makes a 20" meter. Plus if we are talking about high pressure steam we have high heat issues that may affect the ability of an ultrasonic meter to work correctly due to high heat.
An orifice meter in combination with dp, pressure, and temperature transmitters measures mass flow quite well in steam service in combination with good staturated steam tables from good source. Crane Technical Paper No. 14.
Rosemount makes a flow transmitter with the above three integrated into one device. Rosemount 3095 multivariable
flow meter does steam mass flow calculations from ASME steam tables.
On your orifice taps I would make the taps at 45° angle up from horizontal. If you suspect lots of liquid carry over
on your orifice plate I would put a 1/4" weep hole at bottom
of plate to let liquid through. Also I would double the required distance for required pipe run. On a 20" line we are talking long line 30 pipe diameters upstream and 20 pipe diameters down stream. Also put your Thermowell for
RTD beyond 20 diameters downstream. I would install the Flow Transmitter as close as possible to the orifice taps with seal pots to reduce temperature on transmitter.
Hope this helps and Rosemount can give you some guidance with the 3095 Mass flow measurement for steam.
Mark
I have to assume that it is steam line and not super heated.
Coriolis will not work no one makes a 20" meter. Plus if we are talking about high pressure steam we have high heat issues that may affect the ability of an ultrasonic meter to work correctly due to high heat.
An orifice meter in combination with dp, pressure, and temperature transmitters measures mass flow quite well in steam service in combination with good staturated steam tables from good source. Crane Technical Paper No. 14.
Rosemount makes a flow transmitter with the above three integrated into one device. Rosemount 3095 multivariable
flow meter does steam mass flow calculations from ASME steam tables.
On your orifice taps I would make the taps at 45° angle up from horizontal. If you suspect lots of liquid carry over
on your orifice plate I would put a 1/4" weep hole at bottom
of plate to let liquid through. Also I would double the required distance for required pipe run. On a 20" line we are talking long line 30 pipe diameters upstream and 20 pipe diameters down stream. Also put your Thermowell for
RTD beyond 20 diameters downstream. I would install the Flow Transmitter as close as possible to the orifice taps with seal pots to reduce temperature on transmitter.
Hope this helps and Rosemount can give you some guidance with the 3095 Mass flow measurement for steam.
Mark
-
1
- #10
Assumptions
Chemical
If one is planning to use an orifice plate then plan to have a means removing the plate for replacement on a regular basis for wet steam. Never mind the problem of two phase flow and weep holes plugging.
steam4
Mechanical
- Jun 21, 2003
- 23
Maybe I am missing something here......why in the world would you leave condensate in the steam line instead of removing it as soon as you possibly could!!
As dbevil stated, you can measure the condensate after draining it.
Also, is there no place to measure steam flow at the source while it is still "dry"?
As dbevil stated, you can measure the condensate after draining it.
Also, is there no place to measure steam flow at the source while it is still "dry"?
- Thread starter
- #12
Steam4
Good point you'r raising.
Simple answer: We have a Geothermal (Natural) field with 8 wells, each produce mix of steam and water. We would like to measure the flow of each well for predicting well maintenance and cleaning. Other point is when we have a desturbance at the main header to know which of the well is responsible and take care of this well only.
See, not everyone got dry and clean steam...
Good point you'r raising.
Simple answer: We have a Geothermal (Natural) field with 8 wells, each produce mix of steam and water. We would like to measure the flow of each well for predicting well maintenance and cleaning. Other point is when we have a desturbance at the main header to know which of the well is responsible and take care of this well only.
See, not everyone got dry and clean steam...
Beware stating absolutes!
I got caught claiming there where no big coriolis meters. Rheonics make at least an 18" i've heard, so maybe bigger. Plus i know who has the technology to make mass meters any size you want and no bigger than an equivalent turbine meter... but try getting them to manufacture!
Solartron ISA make multiphase meters (as do many others) for hydrocarbons and the Dualstream II is a two phase meter for natural gas production. (So if enough people need something, someone will sooner or later solve the problem. In this day and age it may mean that someone already has your solution and is just shy about letting the world know about it!
I got caught claiming there where no big coriolis meters. Rheonics make at least an 18" i've heard, so maybe bigger. Plus i know who has the technology to make mass meters any size you want and no bigger than an equivalent turbine meter... but try getting them to manufacture!
Solartron ISA make multiphase meters (as do many others) for hydrocarbons and the Dualstream II is a two phase meter for natural gas production. (So if enough people need something, someone will sooner or later solve the problem. In this day and age it may mean that someone already has your solution and is just shy about letting the world know about it!
You could try a wire mesh instrument. Rossendorf research institute in Germany (Dr. Prasser) has experience in developing wire mesh two-phase flow meters. A 20" version could be calibrated at Delft Hydraulics Labs over a wide range of void fractions.
Just a quick note, orifice fittings are problematic in two phase flows. They are not permitted under AGA Report Number 3 (API 14.3) as suitable measurement instruments.
I concur with other observations, ultrasonics is perhaps the best bet.
Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada
I concur with other observations, ultrasonics is perhaps the best bet.
Kenneth J Hueston, PEng
Principal
Sturni-Hueston Engineering Inc
Edmonton, Alberta Canada
- Thread starter
- #18
Assumptions
For your question what we do with the liquid phase: We convert the heat into Electricity. Please stop at:
and see the Binary cycle Energy Converter.
For your question what we do with the liquid phase: We convert the heat into Electricity. Please stop at:
and see the Binary cycle Energy Converter.
Assumptions
Chemical
Wasn't thinking of that but if one separates the two phases then they each can be measured.
Alternately an impact device could measure the momentum and etc. etc..
If one is measuring steam and condensate then there is going to be a lot of wear. A U tube device measuring flow will experience the same results.
If one takes a model such as Briggs....or another and uses different size lines with measurements in each line then perhaps an empirical relationship could be developed.
Sure that there is a way but don't know a device capable of doing it, only ideas.
Alternately an impact device could measure the momentum and etc. etc..
If one is measuring steam and condensate then there is going to be a lot of wear. A U tube device measuring flow will experience the same results.
If one takes a model such as Briggs....or another and uses different size lines with measurements in each line then perhaps an empirical relationship could be developed.
Sure that there is a way but don't know a device capable of doing it, only ideas.
You might want to have a read of this paper concerning ultrasonic flow meters for two phase flow. As discussed above, a coriolis flow meter might be an option also. This gives some insight.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Question