• Whirl as you know can be extremely difficult to detect (unlike stick slip, bit bounce, etc.) from surface data. Some say that it cannot really be detected using surface data. So, what is the trick to this model that we have posted here? The trick is that we infer (not detect) a whirl (more accurately – some dysfunction that cannot be explained – but most likely lateral vibration / whirl – because we rule out everything else). Yes, that is the logic that this model uses.

    Now, you all know that lateral vibrations are mostly seen in hard and interbedded formations, during transitions from soft to hard formations, and are characterized by very low depth of cut. An over-gauge hole also increases the possibility of lateral vibrations. With that mini-intro, let us proceed to break down the model.

    Low ROP and increasing MSE generally indicate dysfunction or that we are drilling through hard rock. So, we track these signals. A low torque variation index implies less probability that the dysfunction is stick slip. A constant WOB trend implies less probability of bit bounce. A decreasing depth of cut and a decreasing bit aggressiveness further support the potential for whirl, and the model shown here brings all this together. Check out slide # 3 where this is illustrated.

    So, let us know what you think about this model – the good, the bad and the ugly. Next week, we will finally go into how to use all the models shown to date to optimize your ROP. And if you are interested, we will tell you how ROP optimization relates to boxing, ice hockey and football. So, stay tuned. Follow us. Be kind and share the knowledge. 

    Click below for slides on this topic:

    Whirl Detection