If you want to go the poor man route, you can make your own assuming you have a shaker motor and FFT analyzer.
What we have done in the past was to take the frame of a metal table (made of angle iron) and bolt it to the floor. We then welded a metal pad about 4"x4" in each of the corners at the top of the table. It did not have a solid top - just the frame.
We purchased a 3" thick rubber sheet from McMaster-Carr and cut our own isolation pads to fit the metal pads at the top of the frame. You'd have to play with the density of the rubber to get something sturdy yet semi-absorbant.
We then took a 3/4" thick sheet of plywood and cut it to fit the top of the metal frame. In this wood we drilled our holes for mounting motors in a horizontal position. The shaker motor was mounted to the top side of this board. The test motors were mounted underneath upside down. This board sat on top of the rubber pads. So the board with mounted motors floated on the rubber pads. The shaker motor should be centered on the top side and your test motors should be symmetrical on the bottom side.
Run your power to both the shaker motor and your test motors. Use strain-relief mesh/cords etc.. on your electrical wires. We suspended the shaker motor power line from the ceiling and the test motors were run up from the floor. And of course have the emergency stop button nearby.
You can adjust your shaker motor's counter-weights inside of either endshield cover to get the amount of vibration that you want. Use the FFT analyzer to get the reading of the system to help with the adjustments. If you have a logging system in the analyzer you can set up your pickups on the motors (may need to glue on metal washers if aluminum frame/endshields etc..) and just let it run and record your vibration levels over time.
If you have problems with too much vibration despite adjusting the shaker-motor then look into buying a more dense rubber material or adding some secured weights to the top of the board.
Our table was capable of running 4 motors at a time. We tested both 56 and 140 frame sizes. It wasn't very large but we used it to do accelerated life testing. If you are looking to test precision equipment or your test subjects are very susceptable to vibration, then you should look at purchasing something less crude and more accurate to dial in the vibration levels better. Of course the mini-motors inside of Playstation 2 controllers look ideal to use as shaker motors for testing PC boards.