Extreme value at start of FFT proccesed signal

I have a signal with frequency of 1000 Hz, and it consist of velocities of an hot-wire. The measurements are taken behind a cylinder as air is blown on the cylinder. When i process the signal of each of the cases (different start velocity) I get an extreme value at the start, just a single point. What is the best way to treat this, because it is an abnormality and I should "remove it"?

What is best to do in a case like this, should I use a window function like hanning or is this actual good data?

• Subtract the average and run again. You can also plot the log(amplitude) to have it be less visually jarring. – Byron Wall May 8 at 19:34
• Once you remove the large peak at 0, you will then see that your results are mirrored across 500 Hz. You will presumably want to only plot from 0 to 500 Hz. – Byron Wall May 8 at 19:36