One Week to Go - Flow Visualization
By Fanny Zuniga
January 14, 1998
This week the test team and I are busy trying to solve a new mystery. The balance has been calibrated and it looks good. But when we plugged the balance into our computer we didn't get a very good signal. There are some electrical components between the balance and the computer (like amplifiers and signal conditioners) and we don't know where the problem is yet. The big worry is that we can't do a final check of the balance with our computer until we've solved this problem. Some of our team members from Boeing are showing up here this week, so with the extra help we should still be able to start our test on time. We are supposed to start putting our model in the tunnel on January 20. At that point the model, balance, instrumentation, and software have to be ready.
On another front, the software group chose this week to make a major upgrade to the wind tunnel software. This means the people who were working on my test's particular software are busy. Some of the calculations I want in our software may not be available until after our test starts. Also, we found those accelerometers I mentioned before, but now we can't find the cables and power supplies. If it isn't one thing, it's another.
Lastly, we are also refining our plans and preparations for using those techniques I mentioned before which help us understand how the air flows around the model. You may recall I mentioned one way we were going to do this is with colored oil that flows on our wings while the tunnel runs. The problem with oil is that we can only get a picture of the airflow for a few conditions because it takes so much effort and time. The other methods of "flow visualization" we will use are tuft photos and Pressure Sensitive Paint, so I want to describe them for you. This week we decided when in the test we would use these different methods.
The first method is to put really fine threads, or tufts, all over the model. These tufts are so fine that they don't change the airflow on the model, but they do show which way the air is going. We take a picture in Ultraviolet light (UV, or Black Light) because the tufts are florescent. We have to use high powered UV strobe lights so that we can "freeze" the tufts because they tend to wiggle a bit (the airflow in our tunnel is not perfectly steady) and we don't want a blurry image. So all we have to do is mount special UV strobe lights above a clear window in the top of the test section and take pictures. What's neat about this is we don't have to stop the tunnel and reapply oil for each condition we want to study, if the airflow changes at a new model attitude or tunnel speed, we just take another picture. It takes time to stick the tufts on the model and install the cameras, but after that we can get a lot of information without slowing down our test too much.
The other method we plan to use is to paint the model with Pressure Sensitive Paint, or PSP. This is a neat method of recording pressures on the surface of the wings. Basically, we use a special paint that glows in UV light. How bright it glows depends on the air pressure. We take video images of the painted model using UV lights in the top of the test section, and then process the images with a computer. PSP is a lot like using the tufts, it takes time to paint the model and install the cameras, but after that we don't have to stop the tunnel and repaint it for each condition like we do with the oil. After it is applied, the paint for PSP only lasts for a few weeks before it ages and quits working.
We are using one more method to understand the aerodynamics of our model. You see, one problem we have is that the wings bend, or deform, slightly under high lift loads. What this means is that the shape of our model will actually change slightly with different loads and conditions. This means it is hard to compare Lift and Drag from one condition to another because the model itself has changed. We use a neat video system, mounted in a side window of the tunnel, which records images of some reflective spots on the wing tip and then computes how much the wing is bending and twisting.
So, the big decision we had to make this week was when we would conduct the oil flow study, and when we would paint the model for PSP. Yes, we are still refining our run schedule! We decided to put both of these activities near the end of the test. Basically, we want to put things that are either risky or time consuming near the back of the test so that we can assure we collect our most important data (like Lift and Drag force measurements) first. As I mentioned, the oil flow photos are time consuming. As for PSP, since the paint doesn't last very long we have to paint inside the tunnel in the middle of our test. The tradeoff here is that after we paint the model we have to change the model back to some configurations we will have already tested. If we painted the model during the first week, say, then we could get PSP information anytime during the test until the paint got old. Then we'd have to repaint it which takes more time.