edit: See update in post 2. Had to offset the reference and most of what follows cleared up. Leaving what I discovered along the way for the next person instead of deleting.
You get what you pay for
The external sync feature never worked. On top of that, there was extreme phase jitter and RF leakage with the outputs off. Lifting the power leg on that chip with the red dot (I added the dot) disabled the useless back ports and cleared up much of the phase jitter.
But I wanted to use an external reference, so I hot aired off the 50MHz oscillator chip (that outputs CMOS level RF) and injected the CMOS output from the Leo Bodnar GPSDO. I expected that to make this thing dead-on frequency, but it didn't. To get the outputs close, I have to set the fine tuning adjustment to 57.
Since that fine tune control is doing roughly 5Hz per step, getting it perfect is not possible.
And the frequency is still not super stable, its doing the same thing it did with the internal oscillator after being warmed up, bouncing around a couple Hz. The GPSDO outputs 10MHz on port 1 for my bench counter reference and 50MHz for the new signal gen reference. If I put the 50MHz from ch2 or the output of the rubidium into the FC, they are both dead on and rock solid, but this sig gen running off the confirmed-stable GPSDO is not outputting a stable signal. Set to 27.383996Hz and the fine tune at 57, the output bounces between 27.383999 and 27.384001 with an occasional small excursion beyond that. For comparison, with the fine tune at 50 (center) and the sig gen set to 27.385000, I get 27.384903-27.384905.
You get what you pay for
The external sync feature never worked. On top of that, there was extreme phase jitter and RF leakage with the outputs off. Lifting the power leg on that chip with the red dot (I added the dot) disabled the useless back ports and cleared up much of the phase jitter.
But I wanted to use an external reference, so I hot aired off the 50MHz oscillator chip (that outputs CMOS level RF) and injected the CMOS output from the Leo Bodnar GPSDO. I expected that to make this thing dead-on frequency, but it didn't. To get the outputs close, I have to set the fine tuning adjustment to 57.
Since that fine tune control is doing roughly 5Hz per step, getting it perfect is not possible.
And the frequency is still not super stable, its doing the same thing it did with the internal oscillator after being warmed up, bouncing around a couple Hz. The GPSDO outputs 10MHz on port 1 for my bench counter reference and 50MHz for the new signal gen reference. If I put the 50MHz from ch2 or the output of the rubidium into the FC, they are both dead on and rock solid, but this sig gen running off the confirmed-stable GPSDO is not outputting a stable signal. Set to 27.383996Hz and the fine tune at 57, the output bounces between 27.383999 and 27.384001 with an occasional small excursion beyond that. For comparison, with the fine tune at 50 (center) and the sig gen set to 27.385000, I get 27.384903-27.384905.
Last edited:
