Happy New Year everybody! I wish everyone lots of good things and happiness!
After many tests and modifications, I should admit that I wasn’t happy with the 3 band EQ results – it either sounded wrong or was unstable at some EQ settings. Turned out making a single Baxandal EQ circuit with three independent frequency bands is difficult to design 😂. I always wondered why classic Ampeg amps had their mid frequency control made as a separate stage – now I probably know why!
So I decided to ditch the mids and only keep two bands, bass and treble. Still thinking about making corner frequencies selectable by the switches, two frequencies per band. This could be useful for different instruments, like moving treble cut-off lower for the bass but higher for the guitar…
This is how my finalized prototype looks now with the mid frequency components out:
Its schematics:
And frequency curves (these worked the best with my basses)
My initial idea was to create a circuit inspired from the classic tube bass amplifiers, like Ampeg Portaflex or SVT. I did some research and got somewhat carried away, so the circuit I came up with has ideas from many sources and doesn’t reproduce any of the existing modules. There were also ideas suggested by OnlyBass forum members that I’ll try to test out and maybe incorporate into the design.
So this is the first version that I came up with and ran through LTSpice simulation:
This is basically a “symmetrical Baxandal EQ in the negative feedback circuit”, with RC Bass and Treble and LC Mid sections. The LC/RC bridges are all high-impedance and require high-impedance next stage, hence the usage of the bootstrapped emitter follower (Q1) as a buffer with 250-300K input impedance. Q2 is the identical input buffer, and the Q4 buffers out the output – all of these make the EQ module completely independent and usable all by itself (perhaps as a separate pedal). As is, this EQ can be also easily transplanted into a tube circuit with minor or no modifications.
The pots are linear 100K, easy to source and more precise than logarithmic pots used in some of the vintage passive tone stack designs. For the Mid EQ inductor I’ll be using a primary winding of a small audio transformer, I have some handy but will need to test them in the circuit – it’s unclear what their inductance is, depending on the source it’s from 3 to 11 Hn.
The overall amplification is 0db (no amplification), the frequency bands give about 12 db cut and boost for the Bass and Treble and 8 db cut and boost for the Mids.
One of the things suggested on OnlyBass is to make the band frequencies modifiable, and I will be certainly testing this, to do that I just need to vary the capacitor values: C8 for the mids, C4 for the bass and C16/C21 together for the treble. In the final product I’ll maybe choose two values for each of the bands (using three switches, one per band).
The tests in real hardware have shown very close characteristics to what LTSpice have simulated, this means the models of the transistors I’ve been using are pretty accurate, that’s good news!
i’ll be testing some slight modifications – reducing maximum gain and maximizing undistorted output – smaller R9 and R10 and higher R8.
The first stage (Q1 and Q2) didn’t need any modifications whatsoever. Even the modifications I’ll be doing are pretty minor and may not be necessary, time to try this and see if it worked out better at all!
This preamp / effects pedal isn’t supposed to emulate the sound of Mike Kerr, I’ve just got some ideas I would like to try 🙂
Just for the kicks and as a challenge, I’ll be building this using only (or at least mostly) germanium transistors. I’ve been compulsively buying germaniums for quite some time, need to put them to good use!
