The math behind this EQ circuit using inductors.

[scott2000]

So if I'm thinking about this correctly:

Here's how they did it in the sound shaper3.. Circuit operation in SM is detailed well... Sounds a bit familiar actually...

 

[abbey road d enfer]

witch one?
the serial or the parallel?

Both. Because of the center-tapped pots.
Now I'm not a fan of the parallel approach because of te interaction between bands.
BTW, circuit b does not work, with the outputs of the opamps all in parallels.

 

[thor.zmt]

View attachment 114108

Some 2 years ago, i saw / found a EQ guitar pedal circuit.
I did a new search and found the upper circuit with is pretty similar.

There is a connection missing in this schematic...

Back than, i bought the correct inductors, but they where SMD and pretty small to handle when i got them

I VERY MUCH doubt you got the correct inductors. 1H is invariably big, it needs a pretty chunky RM core with a lot of turns. If you check, you probably got inductors with their value in microhenry, not millihenry.

Others have helped with the calculation, however, I would suggest to look long and hard at the inductor situation.

You can find 10mH, 68mH & 150mH relatively easy, these would be usable for the upper three bands:

mouser - RF Inductors - Leaded

If you use a Quad Op-Amp (TL074 or OPA1679) for this circuit you have two spare Op-Amp's to use "gyrator" for the LF EQ bands where inductors are not so easy to get.

Of course, this then raises the question why not a fully gyrator EQ like so:


Once you use quad Op-Amp's the actual physical parts count is not really larger than the LC EQ and the parts are more easily sourced.

Two TL074 or OPA1679 will allow for the above circuit, plus input buffer and an extra spare Op-Amp that could be used as rail splitter if supplied by a single 9V battery.

Thor

 

[Analog_Fan]

Both. Because of the center-tapped pots.
Now I'm not a fan of the parallel approach because of te interaction between bands.
BTW, circuit b does not work, with the outputs of the opamps all in parallels.

i did a simulation.

serial:
https://tinyurl.com/29a9e5qh
parallel:

https://tinyurl.com/2cmdahgx
the parallel doesn't seam to be good, at some time the "fase" starts lagging or shifting and the opamps "fight" each other and the power consume from the opamp's output goes up, just a few milli amp or twice the consume of the serial approach.

 

[Analog_Fan]

There is a connection missing in this schematic...

View attachment 114217



I VERY MUCH doubt you got the correct inductors. 1H is invariably big, it needs a pretty chunky RM core with a lot of turns. If you check, you probably got inductors with their value in microhenry, not millihenry.

Others have helped with the calculation, however, I would suggest to look long and hard at the inductor situation.

You can find 10mH, 68mH & 150mH relatively easy, these would be usable for the upper three bands:

mouser - RF Inductors - Leaded

If you use a Quad Op-Amp (TL074 or OPA1679) for this circuit you have two spare Op-Amp's to use "gyrator" for the LF EQ bands where inductors are not so easy to get.

Of course, this then raises the question why not a fully gyrator EQ like so:

View attachment 114219
Once you use quad Op-Amp's the actual physical parts count is not really larger than the LC EQ and the parts are more easily sourced.

Two TL074 or OPA1679 will allow for the above circuit, plus input buffer and an extra spare Op-Amp that could be used as rail splitter if supplied by a single 9V battery.

Thor

THNX

That's what i did last night, after reading an answer is another thread.

The gyrator approach, link to Falstad simulation: https://tinyurl.com/236get6g

An audio circuit collection, Part 3
https://www.ti.com/jp/lit/an/slyt134/slyt134.pdf(Texas Instruments documents on build EQ's).

There is also a high pass circuit in there, but doesn't seem successful.

The capacitor are non standard it seams.

2.3E-08 = 23 nF?
4.0E-07 = 400 nF?
5.3E-09 = 5.3 nF or 5300 pF?
6.4E-06 = 6.4 uF?


I did build an EQ a couple of years ago, but i lost the CAD files, the device works, but can't make a front panel for it anymore.

 

[thor.zmt]

The capacitor are non standard it seams.

2.3E-08 = 23 nF?
4.0E-07 = 400 nF?
5.3E-09 = 5.3 nF or 5300 pF?
6.4E-06 = 6.4 uF?

2.3E-08 = 23 nF? -> Use 22nF or 22+1nF if you want to be very accurate.

4.0E-07 = 400 nF? -> Use 390nF or 390nF + 10nF if you want to be very accurate.

5.3E-09 = 5.3 nF or 5300 pF? -> Use 5.1nF or 5.1nF + 220pF OR 4.7nF + 560/680pF if you want to be very accurate.

6.4E-06 = 6.4 uF? -> Yup, use 2.2uF + 2.2uF +1uF + 1uF (as example) to be able to use small size parts.

I did build an EQ a couple of years ago, but i lost the CAD files, the device works, but can't make a front panel for it anymore.
View attachment 114220

Ahhh, PAin in A... Electronics, this is a semi-parametric EQ.

Make the front panel like we did in the old days.

AL stock, brushed & anodised, drill holes, use dry transfer lettering for indications and spray with clear matte lacquer.

Or print everything on a laser printer/copier from PDF onto clear stock and add that as overlay on the front panel. This works especially well if you do everything in negative and wedge it between two layers of clear acrylic can apply LED backlighting.

Thor

 

[Somkereki]

https://en.cppreference.com/w/cpp/language/operator_arithmetic
bitwise XOR a ^ b T operator^(const T &a, const T2 &b);

It can't handle "double values"



https://en.wikipedia.org/wiki/RLC_circuitView attachment 114158

I'm looking at band pass filter, correct?

There are some calculators here:
https://electronicbase.net/band-pass-filter-calculator/
This is the formula of the upper: Opamp Labs Inc <> Cap-Freq-Ind CALCULATOR <> www.opamplabs.com
double test = 159.154/std::sqrt(inductor * capacitor);
(works)


https://electronicbase.net/band-pass-filter-calculator/View attachment 114156

double test = 1/(2.0 * M_PI * std::sqrt(inductor * capacitor));

returns a number that's 1000 x smaller that the above formula.


The wiki formula (from above):
double test = 1/(std::sqrt(inductor * capacitor));
returns a number not even close.

 

[Somkereki]

f you allow me a comment: (I'm also an analog fan) I started quite a long time ago and I'm already 68 years old. I was with "road" rock bands (at mixing consoles). In fact, I designed and built a "full extra" mixing console in 1980, which was also used a few years ago at the Hungarian Rock Theater. I would prefer the inductance solution, even though it is sensitive to stray magnetic fields. If you use an active circuit, you don't have to worry about the nonlinearity of the iron cores, although I remember that I liked the sound of the Akai GX head tape recorders less than that of the permalloy head tape recorders. They are usually made with a ferrite iron core, but I have also found a pleasant sound path among them, such as the Siemens iron cores. If you only get an iron core, you can wind the inductances yourself. If you don't have an accurate L-C meter, you can set them accurately with the resonance frequency. Winding up a little more thread, slowly unwind the threads to the required value. Today's oscilloscopes also accurately show the frequency... I wish you good luck and above all patience. I would be happy to help you in practice, but on September 14th I am going for a coronary artery examination, which may be followed by surgery, that is "two chances" I am not nervous, because if call the Creator, kept me until now...

 

[gyraf]

..good luck with the procedures, @Somkereki..

 

[thor.zmt]

that is "two chances"

Good luck. I am sure you will pull through.

Remember, you are not done on this dirtball, there is a lot left to do, no skiving off mate!

Thor

 

[Analog_Fan]

f you allow me a comment: (I'm also an analog fan) I started quite a long time ago and I'm already 68 years old. I was with "road" rock bands (at mixing consoles). In fact, I designed and built a "full extra" mixing console in 1980, which was also used a few years ago at the Hungarian Rock Theater. I would prefer the inductance solution, even though it is sensitive to stray magnetic fields. If you use an active circuit, you don't have to worry about the nonlinearity of the iron cores, although I remember that I liked the sound of the Akai GX head tape recorders less than that of the permalloy head tape recorders. They are usually made with a ferrite iron core, but I have also found a pleasant sound path among them, such as the Siemens iron cores. If you only get an iron core, you can wind the inductances yourself. If you don't have an accurate L-C meter, you can set them accurately with the resonance frequency. Winding up a little more thread, slowly unwind the threads to the required value. Today's oscilloscopes also accurately show the frequency... I wish you good luck and above all patience. I would be happy to help you in practice, but on September 14th I am going for a coronary artery examination, which may be followed by surgery, that is "two chances" I am not nervous, because if call the Creator, kept me until now...

Well, i got time, it will be far in to next year before i can buy a Fluke multi meter that can measure inductance.

Or i could go this route:


my "old" oscilloscope can measure frequency.

It's definitely something on my TODO list to somewhat "master" and make inductors.

Hopefully good news will follow September 14th