Are you just putting the tones in the center of each fft bin, then subtracting the tones’ (known) energy from each bin on analysis ?
When Multitone generates tones, it creates them centered on an FFT bin. These are then filtered using a notch filter when computing TD+N. Multitone can also detect frequencies in an externally generated test signal, or use one that's been saved to a file.
Edit: here's a processed 5 multitone signal. You can see the notched signal in white, full recorded signal in blue:
Last edited: