I've been looking at several drilling rate of penetration (ROP) models and noticed that many of them have empirical factors which are undefined in the original publications. This includes Maurer (1962), Galle and Woods (1963), D-exponent (Bingham; Jorden and Shirley; Rehm and McClendon), Modified Warren (Hareland and Hoberock, 1993), Bourgoyne and Young (1974), Pessier and Fear (1992), etc. methods. Is anyone aware of suitable ROP models and the associated empirical factors for drilling in certain rocks? I'm mostly interested in mudstones and sandstones, though most methods are calibrated in shale. It seems most methods derive such factors from adjacent wells in the same field, though I have no such luxury as existing data to calibrate with, so some idea of indicative values would be useful.
Regarding the Bourgoyne and Young (1974) method, the a[sub]1[/sub] (drillability) parameter seems very ill-defined and not related to the actual strength. This is a method where weighting factors are derived from multiple regression analyses, so it's kind of a massive field-specific fudge approach. Does anyone know of weighted factors which are related to real physical properties?
Related to the Galle and Woods (1963) method, the bit wear calculations seem odd, as far as I can tell. You need to define the initial bit wear to calculate factors which are subsequently used to calculate the rate of change in bit wear. If you have a fresh bit with zero wear, the calculations are messed up as one of the factors becomes divided by zero. Results are also highly sensitive to this initial estimate of bit wear. Perhaps I'm going about this in the wrong way, but it doesn't seem a particularly good bit wear model insomuch as you have to essentially predict the depth at which you'd get a certain amount of wear, rather than being able to incrementally estimate it. Has anyone encountered similar issues and applied a better bit wear calculation approach? If so, what bit wear model was that?
In addition, I'm getting some very odd results when using the Pessier and Fear (1992) approach for drilling at small diameters - drill rates of apparently 39,000+ m/h, which is unbelievable. This will of course be limited by bit hydraulics, but it's disappointing that the method is so far out when drilling at small diameters. I wonder if this is because my confined compressive strength (CCS) is relatively low compared to conventional (deep well) applications: water depth at the site is as low as 20m with a further 35m of drilling, and we'll be drilling with a low density mud, so there is essentially 55m + pump pressure + negligible contribution of fluid dynamics of extra confining pressure, so the CCS isn't that much greater than the unconfined compressive strength (UCS). Has anyone encountered similar issues or developed ideas of a better approach?
Regarding the Bourgoyne and Young (1974) method, the a[sub]1[/sub] (drillability) parameter seems very ill-defined and not related to the actual strength. This is a method where weighting factors are derived from multiple regression analyses, so it's kind of a massive field-specific fudge approach. Does anyone know of weighted factors which are related to real physical properties?
Related to the Galle and Woods (1963) method, the bit wear calculations seem odd, as far as I can tell. You need to define the initial bit wear to calculate factors which are subsequently used to calculate the rate of change in bit wear. If you have a fresh bit with zero wear, the calculations are messed up as one of the factors becomes divided by zero. Results are also highly sensitive to this initial estimate of bit wear. Perhaps I'm going about this in the wrong way, but it doesn't seem a particularly good bit wear model insomuch as you have to essentially predict the depth at which you'd get a certain amount of wear, rather than being able to incrementally estimate it. Has anyone encountered similar issues and applied a better bit wear calculation approach? If so, what bit wear model was that?
In addition, I'm getting some very odd results when using the Pessier and Fear (1992) approach for drilling at small diameters - drill rates of apparently 39,000+ m/h, which is unbelievable. This will of course be limited by bit hydraulics, but it's disappointing that the method is so far out when drilling at small diameters. I wonder if this is because my confined compressive strength (CCS) is relatively low compared to conventional (deep well) applications: water depth at the site is as low as 20m with a further 35m of drilling, and we'll be drilling with a low density mud, so there is essentially 55m + pump pressure + negligible contribution of fluid dynamics of extra confining pressure, so the CCS isn't that much greater than the unconfined compressive strength (UCS). Has anyone encountered similar issues or developed ideas of a better approach?
