| Electronics > Projects, Designs, and Technical Stuff |
| Graphic Equalizer Project. |
| << < (4/5) > >> |
| boB:
Here is a good little article on Q and graphic EQ design... https://www.rane.com/note101.html |
| Audioguru:
The multiple feedback bandpass filters your circuit has show a bandwidth too narrow near its peak and a bandwidth too wide farther down from the peak. |
| Ronaldo95163:
--- Quote from: Audioguru on September 14, 2018, 03:30:02 am ---The multiple feedback bandpass filters your circuit has show a bandwidth too narrow near its peak and a bandwidth too wide farther down from the peak. --- End quote --- Yes this is what i'd like to design for, however I still can't really find anything on selecting Q for a 2/3 octave EQ. I've read the links posted above and they mostly covered 1/2 and 1/3 octave...one mentioned 2/3 I think but didnt say anything on a Q value based on the ISO 266 standard. Q is 4 for a 1/3 octave EQ based on the ISO 266 standard btw. Also what simulation software is that you used? --- Quote from: Lee Leduc on September 10, 2018, 04:04:08 pm ---Here's some info you may find useful. "Designing with the LMC835 Digital-Controlled Graphic Equalizer" Chip is obsolete but documentation has some good design info on selecting "Q". Application note attached. Elliott Sound Products has about 10 EQ design projects. http://sound.whsites.net/projects-0.htm#equ Rane has a good free tech library of EQ, and other professional audio processing things. https://www.rane.com/library.html#gpm1_1 --- End quote --- Came across this from the first link http://sound.whsites.net/project75.htm His design for boosting and cutting the various frequency bands look interesting but were a bit confusing to me particularly the adding of the signal to the CUT circuit I believe first then adding the result to the BOOST circuit. Also He placed pots on the end of each bandpass filter output, but I don't understand them...are the outputs going into the wipers? Normally you wire up a pot between a voltage source and ground and the wiper gives the output voltage...I just can't seem to wrap my mind around this topology for boosting/cutting. I also have no idea how he determined the size of the pots to give him a specific gain...or is this determined by the previous circuit that he said provided a maximum boost/cut of 12dB My idea was after each bandpass filtering stage to add an inverting amplification stage after it and do the gain adjustments there then sum the outputs of all the amplification stages with a summing amplifier. But if this topology looks much cleaner as it eliminates the amplification stage after each bandpass filter. |
| JS:
Be careful what you call Q in an equalizer, that's a pretty subjective topic (there are many different uses of Q in EQs, not everybody uses the same definition of Q) In fact, in a real design you should probably choose what band widths you want for each band, are they constant at any level or changes depending on boost/cut? If I were making the choice I'd probably align them to be flat at some mid level, between 3dB and 6dB, where most will be used close to and having a smoother response there seems better to me than at maximum points where raerly will get to, even less a few in a row, except maybe for cleaning up low end where you don't expect it to be heared at all but cleanin the stage monitors which can't reproduce properly the LF you are cutting away. I like symmetrical filtering, but that's not the rule, which is usually symmetrical Q (with some definition of it, like taking -3dB at maximum settings) which ends with very different filter shapes for boost and cut. There's more a discussion about this in semiparametric eqs than graph but it could apply or worth looking for someone who doesn't know about the topic and is dealing with all this. JS |
| oPossum:
--- Quote from: Ronaldo95163 on September 15, 2018, 10:20:06 pm ---He placed pots on the end of each bandpass filter output, but I don't understand them...are the outputs going into the wipers? Normally you wire up a pot between a voltage source and ground and the wiper gives the output voltage...I just can't seem to wrap my mind around this topology for boosting/cutting. --- End quote --- The pots distribute the output current of the filters to an in-phase or out-of-phase summing node. Analyze the circuit with the pot centered, full boost, and full cut to understand it better. --- Quote ---I also have no idea how he determined the size of the pots to give him a specific gain...or is this determined by the previous circuit that he said provided a maximum boost/cut of 12dB --- End quote --- U2B has global influence on the boost/cut range. With the resistor values shown the gain is 3.9. That is 11.8 dB. Change the gain there if you want something other than 12 dB. |
| Navigation |
| Message Index |
| Next page |
| Previous page |