UNID 6928 LSB 0220+ UTC 16 March 2017

[R4002]

6925 kHz LSB is extremely loud tonight, and is splattering over to the other side band pretty often.  However, I am able to copy a QSO in what sounds like Spanish on 6928 kHz LSB, with heavy QRM from 6925 kHz LSB.  6928 LSB is weaker, but there is certainly a QSO there.  This is an interesting example of the "peskies" using 3 kHz steps instead of their usual 5 kHz steps.

S2-S3 signal levels.  Just above noise floor with very heavy QRM from 6925 LSB and 6930 LSB.

[TheRelayStation]

3Khz steps might be the norm if the bandwidth of the TX is observed on the FFT at 2.5Khz.
this would make sense to have 3Khz steps assuming their TX & RX bandwidth is 2.5Khz.
ive always wondered what theyre saying though, i hear these "peskies" all the time.

[ChrisSmolinski]

Checking my SDR recording, I noticed a QSO on 6928 USB around 0657z. Possibly Spanish or Portuguese.

[R4002]

I've noticed that a lot of the time there will be both USB and LSB traffic on the same frequency.  As TheRelayStation mentioned, assuming b/w is within reasonable limits, this entirely possible (as is 3 kHz steps vs. 5 kHz steps).  6900 kHz is a great example.  I've logged 6900 LSB and 6900 USB at the same time many many times. 

The other obvious example is 6925 kHz