Comparing Table 3 data to Table 2 data, we find for baseline Test 2 and redesign Test 

 Al, both with no weight added, the impact energy increased a small amount presumably from 

 reduced drill body displacement. This was also the case for Test 1 and Test A3. This seems to 

 validate model prediction. However, this does not hold true when comparing Test 3 to Tests A4 

 or A5. It appears that the shorter cycle length in the later tests contributed to reduced impact 

 energy. 



During data collection for Tests A5 and A6, the drill was observed to operate in two 

 modes. The first mode showed an initial displacement of the drill body, and then resulted in a 

 fast cycle with low force impacts about that displacement. The second mode caused the drill 

 body to rise and fall at low frequency with higher impact force. The pressure from the supply 

 also dropped slightly during the second mode operation indicating an increase in flow rate. 

 During Tests A3, A5, and A6, the initial impact was far higher than subsequent impacts 

 suggesting that the initial impact had a direct effect on mode selection and that Test A3 may also 

 have been a first mode operation. Figure 10 shows the drill body displacement and impact 

 energy for this condition. 



Test A7 results are for the drill operating in the second mode with approximately 100 

 pounds applied to the drill. The larger amplitude force impacts tracked with drill displacement. 

 When the drill body displacement was small, the impact strength was greatest. This confirmed 

 model predictions for minimizing drill body displacement to maximize impacts. Figure 11 

 shows, for Test A7, nearly half of the impacts were at 6 foot-pounds or greater, though there 

 was still a large spread in the range for impact energy. 



At 1,000-psi operating pressure and with 100 pounds weight added to the drill, the 

 redesign model showed slightfy higher impact energy (2.3 to 3.3 foot-pounds versus 1.2 to 3 

 foot-pounds), significantly greater drill body motion (0.27 inches versus 0.01 inches), and shorter 

 cycle time (19 ms versus 24 ms). However, when compared with test results from Test A4, run 

 at 1,500 psi with no added weight, the cycle time and body motion of the redesigned model 

 match more closely. 



Damping Coefficients 



The model behavior relating impact energy to added weight/reduced drill body 

 displacement was evident in the hardware tests; however, model behavior seemed to correlate 

 better with results from higher pressure test conditions. Inconsistent cycling in the redesign model 

 and in hardware tests continued to thwart an exact comparison. An analysis of model parameter 

 for the damping coefficients was conducted to further validate the model to test conditions. 



With the longer plunger remaining in contact with the piston until impact with the anvil, 

 the plunger cutout does not dwell in the "open" position (energizing the kicker port from the 

 supply port). In the baseline drill, the plunger stopped in a position that allowed sufficient time 

 to fully energize the kicker port. When the kicker port is not fully energized, the poppet does 

 not close leaving the plunger chamber pressurized for a secondary impact. This "double impact" 

 occurs with much less energy than the primary impact. Sometimes, a tertiary impact with even 

 less energy than the secondary impact may occur before a new cycle can begin. 



Test results exhibited suspected "double-impact" events at higher operating pressures 

 (e.g., Test 6 at 1,500 psi). It was determined that insufficient damping was included in the hard 

 stops, which explained why double impacting occurred in the DADS model at a lower operating 

 pressure than in the test results. In the redesign model, a uniform value of 2.5 lb (in. /sec) was 

 used for damping. This corresponded to a range of 8 to 28 percent of critical damping 



