Mikhail Naganov

JDS Labs Subjective3, Part 2

As I was unhappy with the imprecision of “Mid” tone control on Subjective3, I decided to get rid of it—I don’t see a point in having it anyway. So I opened my unit, and desoldered VR2 pot away. After that I’ve re-measured the frequency response and channel balance, and it got more even:

The inter-channel difference is ~0.3 dB, which isn’t superb, and I’m thinking how could this be fixed.

I’ve also measured the pot I desoldered, and indeed it’s not very well balanced. At the middle (notched) position it shows:

So if you are soldering Subjective3 yourself, think twice if it’s really needed to put the “Mid” tone control in. At least, measure it first to check how well its channels are balanced.

Another thing that was puzzling me since I saw the partial schematics of Subjective3 here is whether it preserves absolute phase. Clearly, the schematics shows that the filter network is attached to the inverting input of the opamp. And I’ve actually checked whether the polarity is preserved by listening to in-phase stereo pink noise, where left channel was going through Subjective3, and right channel was connected directly, and there is no inversion.

Then I’ve found an explanation of this fact by examining the board and finding another stereo opamp there, near the output connectors. So apparently, it is used to restore the polarity.

And yet another thing I’ve figured out by disassembling my unit is that ready-made units soldered by JDS Labs employ DC source coupling. As it can be seen, C2_L and C2_R capacitors are left out, and their mounting holes are short-circuited. This may be a bit risky if the signal source is not of a good quality.