AMZ logo

True Bypass Measurements

There has been much written and discussed about bypass sytems in pedals, and there are several links to my articles on this topic at the bottom of the page. I decided to use my new real time analyzer software to put some pedals to the test and get real world results to review.

The laptop computer used in these tests is running WinXP and using an M-Audio Quattro external sound interface that was calibrated in TrueRTA with a single loopback cable. As you can see from the graph below, the calibrated response of the basic system is completely flat from 10 Hz to 20k Hz. (sweep mode)



The first test setup was with four pedals that use true bypass switching placed in series. Three of them contained the blue 3PDT footswitches and the other one uses the AMZ relay bypass board. There was no reason these pedals were chosen except that they were the four available next to my workbench when the test was started. Hosa cables were used to jumper the pedals togther - nothing exotic or special. The graph below demonstrates that the response with 4 true bypass pedals in the recording loop is still completely flat! There is maybe a 1px down-tick out at 20k Hz but is very very small and inaudible; probably less than 0.1db. This is probably due the the cable capacitance.



Next, the four true bypass pedals were removed from the signal chain and four Boss pedals were substituted in their places. Two were the DS-1 model and two were SD-1 models and all were brand new and unused. No changes were made to the system except substituting pedals. As you can see, the signal is down 2.5db below the -20db input in the mid-range and the frequency response has started to rolloff at both ends of the spectrum. Compare this response to the true bypass response in the graph above.



The tests were made with only 4 pedals, as this was all the Boss pedals that I had available to test. The signal and frequency response loss is accumulative and would only get worse as more buffered pedals are added to the signal path. For reference, I have combined both test results onto one graph shown below. The signal in-to-out should measure -20 db from 10 Hz to 20k Hz, as it does with the loopback cable and the true bypass signal chain, and you can see the comparative magnitude of the loss with the buffered bypass system.



What would the response look like with 10 buffered/jfet pedals on the board?

I probably should have measured the impact of the jfet switches on the noise floor and/or distortion of the signal, but I'll save that for another day.

True bypass is not the only solution, and this doesn’t mean that buffers are bad - on the contrary, they have important uses. However, stacking a series of buffered pedals with jfet switching can lead to signal degradation… in the end, let your ears be the guide!

More Reading:


©2012 Jack Orman
All Rights Reserved
Not for Posting on Any Web Site

AMZ-FX Home Page       Lab Notebook Main Page       Guitar Effects Blog