I modified the pulser circuit a bit by adding 50 ohm series resistance at the source. This made a marked improvement in the pulse ringing and is quite a beautiful fast 4.5V 750ps pulse for almost no work and zero money (the tinylogic parts were free samples
).
**Edit: It is roughly a 2.5V pulse now because of the addition of the 50ohm source series resistance. Rise time 600ps on Tek7104 1Ghz scope.
The interesting thing, and a reason why I am commenting in this thread, is that the Rigol really does seem to have a bandwidth of 140mhz to maybe even 180mhz if I am not doing anything wrong.
In the first pic I measured the pulse with my 200mhz Agilent MSOX3024A and found a rise time of 1.59ns (limited by the BW of the Agilent, as this pulse is less than 750ps on a 1Ghz scope).
In the second pic, I measured the pulse on the Rigol modified DS1052E firmware 00.02.02 SP2 and found a rise time of 2.512ns (normal mode aquisition, real time sampling, sinx/x)
Lastly, in the third pic, I used the same Rigol scope only this time with 16 averages, equal time sampling, sinx/x and found a rise time of 1.928ns! This calculates using .35/rt to a BW of over 180mhz. And the pulse looks great, almost as good as the Agilent (sadly for 1/15 the cost)
Maybe I will keep firmware 00.02.02 after all