Half-way point in the test
- two weeks to go!

February 12, 1998

Off to a slow start but we're finally rolling. We are behind and everyone's thinking about how to make up for lost time.

Monday: We have completed 70 runs to check accuracy, repeatability, sample time for each data point, and agreement with previous test results. We are done with the checkout phase of the test! We are finally ready to change the model to the first takeoff configuration and start doing some research.

On another front, some parts of the minituft imaging system are broken, including the strobe lights and the lens on the camera. We are still trying to fix it, but we may just have to give up on the tuft images for this test.

Tuesday: We had our share of problems today. We got dumped on by "El Nino"; roads were flooded and making it hard for people to get to work. So, this morning the model was ready to run, but we did not have a full crew to run the tunnel. Even the weather is slowing us down. Then, part of the tunnel controls quit working. We didn't get to run the tunnel until near lunch. This sort of stuff not only slows us down, but gets on our nerves. This job can be interesting and frustrating at the same time.

After lunch tunnel problems and software problems kept us from running until early evening. Basically, we are in the part of our test where we think we are getting good data but now are faced with various mechanical and software problems that keep us from running the tunnel. We are hoping to start the last phase of test soon. That's the phase where we should be very productive and get most of our useful results. When everything is working well, we can accomplish a lot in a little time.

Wednesday: Halfway point of the test - at least on the calendar! We started with 5 weeks (25 days), minus 2 days for holidays that occur during our test. That left 23 days. Remember, we are running two full work shifts each day, so we had a total of 46 work shifts for the entire test. We started late 2 days, that's 42 shifts total. Those two days cost us 8 percent of our test!!! Including today we have 24 shifts left for the rest of the test. Then the checkout phase we just finished took longer than planned, so we're more behind. We did plan extra time into our schedule for problems (we call this "contingency" time). But we don't like being this far behind this early, because that means we've used up most of our contingency time before we really even got started.

We are going to have to really crank it up! We decided that, starting Thursday we would work two 10-hour shifts (instead of 8-hour shifts) to make up for some of our lost time. This means day crew runs from 6:00 am to 4:00 p.m. and night crew runs from then until 2:00 a.m. the next day. Ouch! We also will run on Saturdays if enough of the crew volunteer. Ouch, Ouch! The good news is that this will only last a few weeks, sort of like cramming for a final exam. Another way to make up for lost time is to extend the test past February 20 if we can get extra time in the tunnel. This could happen if it's OK with the next test to use the tunnel. Unfortunately, our model is scheduled for use in another tunnel (at Langley Research Center in Virginia). So even if we could get more time in the tunnel, we have to pull the model out after February 24, so we could only get two extra days at the most.

We are really starting to make progress in getting our research runs made, so let me explain a little more of what were looking for. Some of the runs we are making right now have involved changing the angle of the flaps at the front (leading edge) and back (trailing edge) of the wing. I can't show you pictures of this, but imagine the front and back of the wing droop downward. We say that the flaps are "deflected." We basically check dozens of combinations of leading edge and trailing edge flap deflection. Deflecting the flaps makes the wing act like it has more curvature. The result is that both Lift and Drag go up, but not always in the same amounts. We are looking for combinations that work well for takeoff and landing. For takeoff, since the airplane is trying to speed up and climb, we want lots of lift without increasing drag too much. On the other hand, for landing we want to fly as slow as possible; we need all the Lift we can get. Because we are slowing down and descending we don't care very much if there is a lot of Drag. So we can look for a combination of flap deflections that gives us higher Lift than we would use for takeoff. Not only are we looking at flap deflections, we have several different shapes or types of flaps to try. This will give us plenty to do.

We also want data with the horizontal tail on and off. We do this so we can determine the effect the tail has on the model's aerodynamics. We can then figure out how much the wing is contributing to the overall Lift.

Thursday: At this point we are making better progress, but we lost several hours yesterday and again today to software problems. We are getting nervous about finishing everything, so we are starting to look for things to cut out of the schedule. We might drop some of the repeat runs that I described earlier because we have more confidence in our data and we need to save time. In testing, we are always trading off speed (number of runs) and risk (confidence in the data quality).

Good news is that we have enough volunteers to run an 8-hour shift this coming Saturday. That should really help.

Friday: As of this morning we have about 295 runs remaining and 20 shifts left, not including weekends. We have decided to toss out any runs that were just for imaging the mini tufts since we've pretty much given up on getting the system to work. Since we are behind schedule and the balance has been behaving so well, we finally made the decision to cut most of the repeat runs.

Saturday: There was some excitement this morning. We thought we had lost a bunch of our model parts. That would really mess up our test plans. It turned out that all the parts are here. We had one part labeled incorrectly, which messed up the system we use to identify all the parts.

We just passed run # 200! We made 21 runs in 8 hours! Now we're cruising!

	Taking off the horizontal tail.
In this second picture, the tail is still connected to the model by the pressure tubes which you can barely see.
	This picture shows those pesky mini-tufts. They are finer than your hair, and look like fuzz on the upper surface of the wing. When the wind is on, they lay flat and point out the direction of the airflow. Too bad the system isn't working now so I can't show you how they will look during the test.(I bet you can imagine that though!)