Possible Relay Overload

I'm currently using the following opto-isolated relay board: URL

Everything is working correctly, I can switch on and off a small light without any problems. Now as soon as I plug in a larger load like a water pump (120V 1.2A) it flicks on and quickly off. Now I would expect this if the relays were under powered but they are 120V / 10A relays. Am I missing something, is there something else in these board designs that are limiting me?

Any help is much appreciated.

Updated:

Solution: I have stumbled on to the problem trying your guy’s battery suggestion. I have a shift register that is consolidating the input pins to the relay board. When I was hooking up the battery I decided to bypass the shift register too. Low and behold it works both battery and direct power. It seems the shift register is causing all kinds of problems with the inputs, the worst being it is terrible inconsistent (Fails to shutoff sometimes). I will create another question targeted around the shift registers once I know more. Thanks for you help guys.

Answer

The switching of the pump causes much EMI (ElectroMagnetic Interference). This can be conducted, like Wouter suggests, and then an easy way to test is indeed temporarily use a battery, so that the interference can't come in through the power supply. If it works OK on a battery you'll have to decouple the power supply better. Filters! You can use a mains EMI filter, like this one, built into the socket:

Also decouple input and output of the power supply with capacitors, and the power connections of the microcontroller and other ICs. Use 10 µF parallel with 100 nF, as close as possible to the power supply pins.

If the battery doesn't help then the EMI is radiated. Also here decoupling with capacitors should be the solution.