“We are looking for a new microwave frequency source for our experiment with ultracold Rydberg atoms of Rubidium, tunable around 6GHz (to address an atomic transition) and capable of doing frequency ramps of 1 MHz/ms or faster after being triggered by an external signal.”
WINDFREAK TECH RESPONSE:
“The short story is I think we can get you going with something you can use. Maybe with just a little bit of Engineering money.. The long story follows. As of August 2015 firmware, there are currently two ways to frequency sweep on the SynthHD: The fastest is an FM chirp. FM chirp steps in about 20uS per step but is band limited to 10MHz of total sweep range (at 6.8GHz). The number of steps and step rate are controllable. Transitions between steps are phase linear in nature and slew defined by the hardware loop filter of the PLL and a small amount by a software charge pump current setting. You cannot currently trigger the chirp externally. This is possible with some firmware modifications $$. You can currently set up delays between the chirps where
the RF is off for a programmable amount of time.
Larger sweeps step slower, but can currently be controlled by the trigger. You can trigger a sweep, or trigger steps in the sweep. You can also trigger a list mode table of frequencies up to 100 discrete frequencies and amplitudes. The step time for this type of sweep is 4mS. I am currently working on new firmware that will reduce this step time to 100uS (give or take). This might be your best bet and if 100uS per step is ok, costs you nothing but a short wait. I’m planning on being done early next month. Pulsed (on/off) switching time at 6.8GHz is about 20uS. Pulsed switching time below 4GHz is less than 1uS. This has to do with RF on/off isolation problems as frequency is increased. If 20uS isn’t ok, then you might need a diode type spst.
The software manual gives some insight into the power of the hardware and current firmware.