Well testing

[wlka]

1.Does anyone knows what is the best way or the most accurate way to test the production of an individual well?

2.What is the most accurate way to take the watercut percentage of an individual well?

wlka

 

[Fairclough]

The most accurate way to test an individual well will always be to see what comes out of it. Which means to flow into a tank for a day and then let it settle and separate, strap the tank and se what it made. You will have to make adjustments to account for differences in backpressure because a well will produce diffrently depending on the backpressure on the wellhead. Since this is not practical, a portable test separator can be used. They are a bit of a circus but will do a partial separation o the gas and monitor the oil and water emulsion flowing out giving you a water cut figure. Be sure to avoid coriolis type watercut metering if the gas cannot be completely separated or the water cut is too high.
A multiphase meter is probably the easiest way because you do not have to rely on quality of separation.
Depending on the type of flow ou have, viscosities, GVF, salinity changes in the water etc. there is a right multiphase flow meter for you.
It allows for well tests that can last as little as 20 minutes. Be warned that all MPFM's produce pressure drops and you have to make corrections due to the higher backpressure.

AF

 

[d23]

A somewhat accurate test will require that you flow the well for a period of time before worrying about flow rates or water cuts. The well must reach a stable flowing condition before your test data is any value. I work for a major service company that rents well test equipment on 30 day bases. What you will receive for your testing investment is a test that will offer several operational data points during the course of the 30 day test that will allow you to model your production zone.

Most wells (zones) in my area are water drive zones that are very pridictable using a modified Vogle correlation where the water is a stright line and the oil is Vogle (IPR.) For wells in this area the harder you pump them the lower the oil cut gets.

Any large service company like Slb, Baker, Weatherford etc. can do your testing for you.

Hope this helps.

David

 

[soel97]

dear wlka,
to measure the accurate water cut, you could use the MultiPhase FlowMeter, which is using radiation and conudctivity meter to know the amount of water contained in the oil produced. in the water, there are salts dissolved that make the water has a conductivity, which differ from oil measured. But the problem is, we need to continuously maesure the conductivity of standard water wasted from the WTU in our production sites and make it as reference for the water conduictivity in MPFM program.

 

[rguill5]

why not just get the cut from the wellhead sample & spin it out?

 

[40410]

Fairclough


You are suggesting to avoid coriolis flowmeter when the water cut is high and gas/ oil is not separated

what is the water cut above which it is categorized as high

Once a separator is installed to separate the oil/water and gas the coriolis flowmeter should not be a problem.

in onshore coriolis flowmeter is the preferred one as far as i am aware

 

[dhayes]

I don't know your situation, but the most accurate way is a 3 phase test separator, operating at the same pressure as the main separator. Individual meters on the test separator determine the volume produced during the test period of gas, oil, and water. If the test separator operates at a different pressure, the test will not be as useable to allocate the total production between the various wells.
If this is a situation where you can't have a test separator and the associated switching manifold, then the inline 3 phase meters are your best bet. They can be fairly accurate, given the right conditions and the ability to calibrate for those conditions.

 

[KernOily]

Need more info. Is this heavy or light oil? What is the GOR? What is flowing temperature? How much sand in the production stream? Do you need allocation? Are you testing one well, or several? All these need to be answered before you get a decent answer.

The 3-phase meters are useless in TEOR heavy oil production.

A wellhead jar cut/jar test is only a snapshot. In my experience most wells on rod pump do not produce a nice even cut; rather, the flow is in crude and water slugs.

If you can provide some answers to the above questions I can steer you in the (generally) right direction. Pete Thanks!
Pete

 

[kahalikk]

What is the practical diffrence between 2 phase and 3 phase separators? Is it the retention time and the exit points in the separators? Can we measure the liquid rate in a 2 phase separator and water cut by grab sampling at the well-head and centrifuging it and then find Oil rate from total liquid measurement? Will it cause any error?

 

[nonAPI]

wlka,

There are so many different ways to test wells and the "best" way is to tailor it to fit the individual well coditions. I agree with 74Elsinore, we could use more details on the well and its production types, even the area it is in.

kahalikk,

A 2 phase separator separates production into a liquid (water/oil) and gas streams. A 3 phase separator separates production into water, oil and gas streams. Sampling the water cut as you have mentioned can be fairly accurate in certain high rate instances but in low rate instances is generally inaccurate.

Cliff

 

[ami945]

To get a representative test be sure to review well head data and automation data if existing, and confirm no change in production facilities or gas lift parameters if gas lifted well, confirm stabilization for well at least 6 hrs after switch to test line, record any change in wellhead parameters between production and testing line for modelling and adjustment if required. test period is important and should be not less than 12-24 hrs. type of testing tool is important and conventional test separator is consider best one compared with MPFM, AGAR, PHASE TESTER. caliperation is required between one of these and test separator before using it. reviewing well test data with modelling program as prosper or flosys will detect if test data accepted or not and prediction table including senstivity analysis for different reservoir pressure, watercut and gas injection rate will help production engineer while testing in future. for watercut sample be sure that sample point not at 6 clock position in pipeline and frequently watercut samples at different time per day is prefer.

AMI