One of the problems of using an Orion drive or any pulsed based drive, is that the occupants will feel a sudden jolt of force instead of a smooth constant acceleration for artificial gravity. Is there a way of alleviating this problem for pulsed drive ships?
You need to realize that the ship is not accelerated by the atomic blast, it is accelerated by the springs(*) that exist between the push plate and the ship. I.e. while the push plate experiences insane g-forces, the momentum of the push plate is transferred to the ship continuously over an extended period of time. If you detonate the next charge before the springs have pushed the plate all the way back down, you will have uninterrupted gravity on your ship.
Of course, the further forward the push plate is, the higher the force transmitted by the springs will be for virtually any spring design. However, if you give the springs a sufficient preload, you can limit the transmitted force to a pretty narrow range, and thus provide for a smooth ride. How much you are willing to preload the springs is simply a matter of compromise between a smooth ride and beefiness of the springs structure. The higher the preload, the more energy is already stored in the extended springs, and thus the heavier the springs need to be, eating into the the payload capacity. Nevertheless, continuous acceleration of the ship is always possible by using a high enough blast frequency.
*) With a very general definition of "springs": They could be anything from large pneumatic cylinders to electric systems. The point is, that they continuously act to accelerate the push plate downwards.
Wikipedia's long article says:
The pusher plate would be mounted on large two-stage shock absorbers that would smoothly transmit acceleration to the rest of the spacecraft.
A diagram there confirms two shock absorbers, but lacks hard numbers about stroke length, energy dissipation, time constants, variation of acceleration transmitted, etc.
Dan suggested lots of pulses per second. Using lasers to create fusion in pellets might offer a route to large numbers of small pulses through thermonuclear fusion. Do lots of tiny 'bombs' still count as an Orion drive?