III. Carbon Composition Resistor Mod

The prevailing wisdom is that using carbon composition resistors in the audio path instead of the more common carbon film type will produce a "warmer" sound. I replaced three of the stock 1K resistors in the audio path of the TS7 with carbon composition resistors of similar value:

Here are the resistor locations shown on the TS7 schematic:

R1 R30 R49




And here they are as shown on the general Tube Screamer Schematic from the Tube Screamer Technology Page:





Here is what the board looked like before and after the modification:

Before After

TS7 with the JRC4558 op-amp and the output resistor mod.

TS7 with the JRC4558 op-amp, the output resistor mod, AND R1, R30, and R49 changed to carbon composition resistors.





Here are the measured values of the resistors before and after the modification:

Designator Stock value Carbon Composition
Replacement Value
R1 988 ohms 991 ohms
R30 992 ohms 992 ohms
R49 987 ohms 986 ohms


In each section below, one of the recordings was done with the output resistor mod only, and the other was done with the both the output resistor mod and the carbon composition resistor mod. They have been mixed randomly. Select the recording you think has the carbon composition resistors.

Section 1: Blues Riff Recording A (2765K wav) Recording B (2754K wav)
Section 2: Long Power Chord Recording A (1797K wav) Recording B (1904K wav)
Section 3: Short Power Chord Recording A (208K wav) Recording B (231K wav)
Section 4: Lead with Dynamics Recording A (2903K wav) Recording B (2882K wav)
Section 5: Long Single Note Recording A (1064K wav) Recording B (954K wav)
Section 6: Short Single Note Recording A (179K wav) Recording B (172K wav)
Section 7: Long Fourth Recording A (1015K wav) Recording B (1325K wav)
Section 8: Short Fourth Recording A (182K wav) Recording B (134K wav)

Cast your votes via email

Results

The average score so far is 3/8 correct. I certainly can't hear any difference in the recordings, so I'm inclined to think that these resistor changes are having no effect. But someone could prove me wrong...