Julian,
I might be the best person to answer this. Yes, a good noise source will work, but it will suffer from:
1) Dynamic range. Since your energy is spread through the entire spectrum, the energy at each frequency bin is quite a bit lower. You might only get 40 dB of dynamic range, instead of 90 ish.
2) Nonlinearities. If you are trying to test the frequency response of a device, and “zero out” the response at some maximum amplitude for best signal-to-noise ratio, you may have significant intermodulation products affecting your readings. When you insert your DUT and filter some of them out, your flatness may be less than ideal.
3) Lack of automated software support. With the TG44AA / TG124A, you get software specifically written to do the task, saving you time and potential errors.
Uploaded Failed |  |