A sample to surface can be achieved either via MDT sampling as...

A sample to surface can be achieved either via MDT sampling as we are trying to do now on via TLC (drill pipe conveyed logging) or by flowing a well during a DST (drill stem test). The MDT or its equivalent has always been the priority because it can provide the detail required for building a dynamic simulation model which engineers use to predict or forecast oil and gas production. The MDT, in principle, has the ability to gather many pressure measurements in order to determine the gas condensate pressure gradient as the CGR decreases with depth and determine oil and water gradients. Ultimately, the Intersection of the fluid gradients will enable them to determine the location of a free water level (FWL). Having a hydrocarbon fluid sample at surface provides confirmation that the hydrocarbons are mobile. So far, the condition of the hole looks good, therefore they have decided to continue with MDT sampling via TLC, but it is a very slow process given the time reqd to RIH, POOH and move between stations.

If they were to run to the bottom of the well with the MDT on drill pipe, there is a risk of getting the pipe stuck due to differential sticking because the mud overbalance is greater as depth increases. It was mentioned that within about 50m of TD, the pore pressure in the formation appeared to be increasing, which might be a sign of formation overpressure, and therefore the mud overbalance was much less to the point of almost being "on balance". However, it is not worth the risk sampling that deep because generally the deeper you go the greater the risk of sticking due to the increase in mud overbalance.

Well testing could provide the necessary data, but it would not provide the highest level of detail in the pressure gradients within each zone. Nevertheless, if a zone were to be produced at surface the combined samples from the separator and a single phase down hole sample would provide an equivalent pressure gradient within a sand and a gradient for gas vs liquid condensate within a sand. It would also provide the permeability height product, skin factor, initial pressure and possible if flowing for long enough, the distance to the nearest fault boundary.

A well test programme might include separate tests of oil bearing layers from rich gas condensate layers, separate tests for an intermediate CGR layer vs low CGR layer and i also in the case of logs indicating liquid condensate beneath gas. Combining layers during flow testing is probably NOT the best solution if there are no MDT samples because when you combine layers the PVT properties will obviously be combined, and that will lead to a compromised, uncertain outcome in the simulation work.

Regarding the latest attempt at obtaining an MDT sample, it was unclear how long the tool had remained static trying to pump out mud filtrate? All we saw was a colourful CFA display showing evidence of hydrocarbons, but no time scale. However, prior to that they would have spent time pumping out the filtrate. It would be good to know how much time was spent pumping out prior to the component failure. Could it have been due to the temperature rating of the components, or did it just give up the ghost trying to pump out for too long? What us the max temperature rating? By going into things in more depth, it should be possible to make a clear assessment in orde to mitigate further downtime.