25/07/2013

Diamond Bass Compressor


I have read a lot of great things about this compressor.


This is the schematic traced by WhiteKeyHole on FSB.




The bass version has a switch to select the pivot frequency of the tilt EQ between 900Hz and 250Hz. There are no gut shots of it and I couldn't find the schematic or the component changes needed to build the bass version. The only reference I could find was from John Kallas at the Sabrotone post: (link)
I just raised the values of the 100n’s C7, C8,& C18 to 470n’s and slightly changed the frequency cap on the tilt eq.
From what I know, the only differences between the guitar and bass versions could be the changes sugested by John Kallas to eliminate the low end loss and the switch for the tilt center frequency, so I just needed to find how to configure the EQ.

The Tilt EQ


I found and article by Chu Moy on headwize.com (Designing A Pocket Equalizer For Headphone Listening) that explains the history of the Tilt EQ topology.


The guitar version of the Diamond Compressor use this exact configuration and component values (after C2) to get a pivot frequency of 900Hz with a 6db boost or cut on both sides of the tilt.


The only thing left was to find the component values to change the center frequency. I found a great article by Francesco BalenaImplement an audio-frequency tilt-equalizer filter; where he derives the transfer function for this eq design.


Using the next formula we can calculate the pivot frequency FP:


Replacing the component values of the Diamond we get what is expexted:


This is the center pivot frequency for the guitar version:


Deriving ML and MH from the next formulas we can obtain the maximum low-boost asymptote as ML (bass boost) and the maximum high-boost asymptote as MH (treble boost).
We get this:


You take 20 times the log of Mand Mto get the response in decibels:



Wich yields a 6db boost and cut as advertised by Diamond.

I have simulated this filter in LTSPice



And we get the expected result



Note: I have used 47K for RF in my calcs as it is the prefered value.

Changing the pivot frequency


We can change the pivot frequency by changing the value of C. Using 20nF capacitors we get an FP of 251.37Hz, as you can see in the sim plot





To implement the switch as in the comercial version we can use a DPDT to select between two pairs of capacitors:


I have yet to build it, but I learned a lot studing this.

Configure your tilt EQ


I have implemented an on-line Tilt EQ Calulator to help you configure your own filter. Or you can use the next table if you want 6db bass and treble boost and cut.

P1=50k, R=15k, RF=47k
 C Pivot Frequency
 20 nF  251,37 Hz
 18 nF  279,30 Hz
 15 nF  335,16 Hz
 12 nF  418,94 Hz
 10 nF  502,73 Hz
 8 nF  628,42 Hz
 6,8 nF  739,31 Hz
 5,6 nF  897,74 Hz

You can even achieve an asymetric bass and treble boost. The next sim plot is what you get with: P1=50k, R=33k, RF=47k, C=6.8nF



Special thanks to:
Francesco Balena
WhiteKeyHole





3 comments:

  1. thx for the article, I am studyng myself the tilt ckt these days, not on the math parts, which is greek to me, but building and trying different things. now I got it quite working but it may be better. I got a buffer in front of it using one op amp of a dual ic. the 15k are 10k and the 47k are 2.2k in my snippet. used 47n in it and a b25k pot. no r12 and 13.
    I am gonna try your version, using the same buffer in front. Renegadrian

    ReplyDelete
  2. I traced the entire bass version. There are a few more differences in the bass version, and guitar version won't go much lower than 60hz. http://freestompboxes.org/viewtopic.php?f=7&t=13268&start=40

    ReplyDelete
  3. I exactly got what you mean, thanks for posting. And, I am too much happy to find this website on the world of Google. Diamond FX Trading

    ReplyDelete