"Noiseless" input stage

[bool]

Pre-de-emphasis in bass electronics is usually used to optimize the eq stage (a sort of baxandall in most cases) noise which can be significant in case of "overdone" eq ranges (like +- 15 or 18db boost etc). The first stage "boost" or gain, and related noise injection is in most cases miniscule compared to what happens in eq stage.

A 06x was in most real-life cases good enough for serious studio use.

[KarenColumbo]

Just had to post this: 100k Rev log stereo: https://www.musikding.de/Alpha-Potentiometer-16mm-Stereo-100k-rev-log

[Rob Strand]

A 06x was in most real-life cases good enough for serious studio use.

Sure.   I supposed there's two ends of the scale:  At one end is what you can get away with, that would line-up quite well with the TL06x's.   At the other end of the scale is to do the best you can do, that's more along the lines of what FancyLime wants to do.

For the last case you need to consider everything.   Even though the pickup dominates the noise it's easy to make things worse.  Like suppose we say the pickup is 20nV/rtHz and we don't want to degrade it by more than 1dB.   In rough terms the entire circuit cannot contribute more than about 9nV/rtHz.   If we consider only the first stage and the opamp is somewhere in the 4nV/rtHz to 6nV/rtHz zone (ie. quite low noise) then both input protection and feedback resistors cannot contribute more than 7nV/rtHz to 8nV/rtHz.    A single 3k3 resistor contributes 7.4nV/rtHz.   So with care it is possible to get close to the "do the best you can do" case.

Some of the bass preamps use quite high resistor values which make the noise higher than the TL06x.

FYI:  Quite a few of the higher end basses use LF442's which are low power but have better noise performance than the TL06x's.  There's probably about 10 different low power opamps I've seen in basses.  Some people have complained about the LM4250's being too noisy.

[amptramp]

One thing to consider is that the resistive noise depends on the effective resistance of the pickup in parallel with the source impedance.  If you have a 10 K resistance in the pickup in parallel with the input resistor which may be 1 megohm.  If the capacitive reactance is such that pickup is isolated, the 1 megohm resistor will have a thermal noise of 127 nV / root Hz from the 1 megohm resistor.  If the parallel combination of the 10 K pickup and the input resistor is 1megohm, then you have a parallel resistance of 9901 ohms and the noise level is 12.6 nV / root Hz.

Moral of the story is pick the largest coupling capacitor that will guarantee negligible loss at the lowest frequencies of interest.  Do the bandlimiting in a later stage if you want to cut the bass.

If you have a volume control on the guitar, keep it at maximum or you will be adding additional series resistance with additional parallel resistance to ground.  If you need a volume control on the guitar, the low noise solution is to put the preamp ahead of it in the guitar.

[bool]

You can try to connect pu's directly into a opa134 (DC connection, no "input" caps etc) but this will naturally inject all the crud an subsonics into the poor opamp.

You can lpf right at the input, that's easy. 12-15kHz pass-band is all you need.

You can "hpf" within the opamp nfb. 5Hz corner gets rid of most of string flapping and byproducts of being too heavy with the pick or slapping too hard. It makes sense to aim at 6-12dB of gain here, but that's of course TBD wrt your aims. The subsonics will get thru at 0dB (no gain/no loss); one thing to be aware of.

With this, you can get suprisingly close to noise performance of a 071 or 081 with shorted inputs (if you do things right), for not too much money.