Archive for March 2, 2011

Welcome to a pump decoupling noise comparison test.  This test will focus on consistency using just one pump, the Swiftech MCP-35X and compare a direct metal contact (no decoupling), the factory neoprene pad, and a chunk of egg-crate.  I was also experimenting with microphone and sound level meter setup because I wasn’t very happy with my previous video results.  It seems having the microphone on the test bench was artificially introducing some vibration type noise into the audio recording and possibly into the sound level meter as well.  This time around I docked my Zoom H1 on top of my Canon T2i which is sitting on a tripod a couple of feet away.  I also place the sound level meter on some cushion to also decouple the sound meter from vibration.  I’m starting to notice that while fans generally have very little vibration, pumps obviously do to the point where the instruments can be affected by those vibrations if instrumental decoupling isn’t done as well.

Also doing a little work in YMEC software to take a more detailed look at frequency response.  Here are the three scenarios with the egg-crate decoupler as the baseline in the middle overlayed each of the other two so you can see the frequency spike differences.  While the neoprene pad helps, it’s not nearly as good as the thicker softer egg-crate foam.  The downside is that any sort of foam like this will trap heat, but this is just comparing some decoupling scenarios.  Decoupling AND cooling the pump base at the same time is a challenge.  I have heard of some people suspending pumps by tubing alone.

Here is that frequency response comparison at 100% PWM.  Zoom H1 microphone is approximately 2′ away mounted to camera on separate tripod.  This does seem to avoid any vibration induced noise on the microphone itself.  Of coarse this is tested in extreme silence to try and sort out any fine details.  Even with the microphone place a couple feet away, using software A-weighting I’m seeing upwards of 6dbA improvement using the neoprene pad over direct metal contact, and around a 12dbA improvement using the egg-crate as a decoupler.  This is fairly substantial and easily perceived difference in this silent testing environment.  No doubt that decoupling makes a noteworthy improvement to pump noise in silent environments.

A quick A-weighted frequency response comparison/overlay

And the video recording, zoom to the very end for just the 100% PWM sound clip comparison.

The pump used in this experiment was sponsored by Gabe from, thanks!