Choices for a simple transparent 5 to 10 band EQ (for vocals)

[Rob Strand]

I do have TL072.  I even have TL082, which I understand is more HiFi than the 072 (and also JFET).  Should I drop the TL082 in there?  In this case, keep polarity with negative towards output?

The TL072 is supposed to have better noise and the TL082 is a generic part (a little cheaper in the old days), in practice they seem close.   It will work no problems.

The polarity of the caps shouldn't matter so much, negative to output is the default.   

[jfrabat]

I will go with TL072 then. 

By the way, I just got the power supply today.  I plugged the breadboard in, but I am getting no signal.  I did not have time to troubleshoot, but I did checked voltages.  I am getting 13.8VAC at input, +8.65VDC at V+, -8.65VDC at V-, 0V at ground.  So those all look good.  I did find it odd that the LED connected to C6 is getting -17.65VDC...  is that normal?

Tonight or tomorrow I will go through it with the audio probe to see what is going on with the signal.

Felipe

PS: I also got my FMR RNC 1773 and the Presonus TubePre V2 that will go before this in the signal chain.  The RNC was supposedly not working; turns out, it will not work with a DC power supply, so I got a good deal on it!  Just get a new $10 power supply, and PRESTO!

Almost ready to start some serious recording!

[Rob Strand]

I am getting 13.8VAC at input, +8.65VDC at V+, -8.65VDC at V-, 0V at ground.  So those all look good.  I did find it odd that the LED connected to C6 is getting -17.65VDC...  is that normal?

The 17.65V looks fine, 16V to 18V is very normal.  The 8.65V looks a bit low.  From experience I would have guessed the supply was about 12V.  However, no need to guess, we can calculate what it should be.

Supply current:

     Icc = 4 x [NE5532 current (both opamps) ] + [TL082 current (both opamp)]

Using typical values from the datasheets,

     Icc = 4 x 8mA  + 2 x 1.4mA  = 34.8mA       ; NE5532 species total current but TL082 specifies one opamp.

Expected voltage drop across 150 ohm,
     
    Vdrop = 150 * 34.8mA  = 5.22V

Expected opamp supply voltage,

   Vcc  =   17.65 - 5.22 = 12.4.V

If we look at the *maximum* current for the NE5532 it is 16mA;  the max on the TL082 won't make a big difference.
Repeating, the calculations we get Icc = 66.8mA, Vdrop = 10V, Vcc = 7.65V

So typically we would have expected 12.4V, which seems very reasonable.  However,
because of opamp tolerances we might see as low as  7.65V.   Your 8.65V is in this range
so the opamps are in spec but they are on the high side of good and causing the
supply voltage to drop.

For a one-off build like this the solution is very simple, just reduce the 150ohms to about 100ohms.
That should give about 12V.

PS: I also got my FMR RNC 1773 and the Presonus TubePre V2 that will go before this in the signal chain.  The RNC was supposedly not working; turns out, it will not work with a DC power supply, so I got a good deal on it!  Just get a new $10 power supply, and PRESTO!

Very cool, sometimes you can be lucky.

[jfrabat]

The 17.65V looks fine, 16V to 18V is very normal.  The 8.65V looks a bit low.  From experience I would have guessed the supply was about 12V.  However, no need to guess, we can calculate what it should be.

Supply current:

     Icc = 4 x [NE5532 current (both opamps) ] + [TL082 current (both opamp)]

Using typical values from the datasheets,

     Icc = 4 x 8mA  + 2 x 1.4mA  = 34.8mA       ; NE5532 species total current but TL082 specifies one opamp.

Expected voltage drop across 150 ohm,
     
    Vdrop = 150 * 34.8mA  = 5.22V

Expected opamp supply voltage,

   Vcc  =   17.65 - 5.22 = 12.4.V

If we look at the *maximum* current for the NE5532 it is 16mA;  the max on the TL082 won't make a big difference.
Repeating, the calculations we get Icc = 66.8mA, Vdrop = 10V, Vcc = 7.65V

So typically we would have expected 12.4V, which seems very reasonable.  However,
because of opamp tolerances we might see as low as  7.65V.   Your 8.65V is in this range
so the opamps are in spec but they are on the high side of good and causing the
supply voltage to drop.

For a one-off build like this the solution is very simple, just reduce the 150ohms to about 100ohms.
That should give about 12V.

I replaced the resistors.  I am now at 12.87V (+ and -).  Should I leave it there or increase resistance slightly to lower voltage closer to 12V?

As for the sound, I still have not had a chance to troubleshoot...

[Rob Strand]

Should I leave it there or increase resistance slightly to lower voltage closer to 12V?

Absolutely fine at 12.87V.

[jfrabat]

Great!  Will try to troubleshoot it tomorrow; we just finished a REALLY FRICKING LONG video conference!  I am exhausted!

[jfrabat]

A couple of updates.

1. The old breadboard was working all the way to the 3rd band; I could control the frequencies up or down, and it was noticeable.  There was an issue in the 4th band, but I decided to redo the whole breadboard on my larger breadboard, as the old one did not have the buffer, and space was so tight, troubleshooting was going to be an issue.

2. Since I had taken out the pedal I was working on in the large breadboard, I moved everything over, and included the buffer. 

This is the new breadboard:



As you can see, I am also using the dual gang pots (B100K) in this iteration, instead of the fixed resistors.  And ignore the Cap number written in the power side of the caps; those were the old numbers...  Oh, and I also changed to 10K the resistors in the last band's Wien Bridge to reach a bit higher fequency.

Here is the current schematic, for quick reference to the numbering:



In this new breadboard a couple of things came about, with which I need some help:

• Is it normal for R27 and R35 to get hot?  Is this something I need to worry about?
• In order to more or less balance V+ and V-, this time I had to use different values in R27 and R35.  My guess is that the switch (which I included) is changing the resistance; does that make sense?  I am now getting +12.03V for V+ and -11.73V at V- (also note that I went with 120 and 91 Ohms in the Schematic, which is showing current values in the breadboard)
• Currently signal is not making it through the system; I have noticed (by using the audio probe and a line level signal) that the low end frequencies get filtered (a little more with each band) if I listen to the sound on Pin 1 of the cut/boost pots.  But once I get to pin 7 of IC1, the sound is gone.
• Unlike the previous BB, my cut/boost pots are not cutting or boosting.  My frequency select pots ARE working, but the cut/boost seems to be "stuck"

I am going to go over the entire BB once again matching it to the schematic.  But I wanted to share where I am currently at.

[Mark Hammer]

I have to ask, since I feel kind of responsible for sucking you into this vortex-of-trouble, when it works does it do what you needed?

[jfrabat]

Yes, it does.  And don't feel bad, I am actually happy to have gone this route.  I am learning, and that is valuable.  Plus, once I get this working, the sense of accomplishment will be even more valuable to me!

[Rob Strand]

A couple of things up front.  Make sure all opamps are getting power, both + and -.   You should check the DC voltages on all opamp pins.  They should all be less than 0.1V, probably less than 20mV.

If the DC voltages are wrong you need to sort that out first.    If the DC voltages are out in an earlier band, say band 1, it will affect the DC voltages from that point on because the stages are DC coupled.

1. The old breadboard was working all the way to the 3rd band; I could control the frequencies up or down, and it was noticeable.  There was an issue in the 4th band,

It's possible your band-4 opamp had a problem and when you moved the circuit that opamp is now on band-1 and stopping everything.

Is it normal for R27 and R35 to get hot?  Is this something I need to worry about?

Yes it is expected to get warm.  Since your circuit is pulling twice the expected current it has made things a little worse.

In order to more or less balance V+ and V-, this time I had to use different values in R27 and R35.  My guess is that the switch (which I included) is changing the resistance; does that make sense?  I am now getting +12.03V for V+ and -11.73V at V- (also note that I went with 120 and 91 Ohms in the Schematic, which is showing current values in the breadboard)

Doing that is fine.

Currently signal is not making it through the system; I have noticed (by using the audio probe and a line level signal) that the low end frequencies get filtered (a little more with each band) if I listen to the sound on Pin 1 of the cut/boost pots.  But once I get to pin 7 of IC1, the sound is gone.

Perhaps check the DC voltages first.  They often show-up problems.

If you have wiring issue then obviously that's going to cause problems and doing a thorough check by eye certainly helps.

When you have a largish circuit like this sometimes it is useful to get one band working at a time.    You do that by wiring the buffer to the input of each EQ stage (don't forget to disconnect the input of the EQ from the previous stage.) then wire the EQ stage output to the output jack.

Unlike the previous BB, my cut/boost pots are not cutting or boosting.  My frequency select pots ARE working, but the cut/boost seems to be "stuck"

See how you go after checking the wiring.   (I can't quite make out the wiring from the photo but when I first looked it looked like pins 1 and 3 are not going to the right place in all cases.)

You can try a few other things:
- disconnect the wipers of the boost/cut pots and see if the circuit comes to life.
  Once you disconnect the boost/cut wipers the circuit should just pass signal.
- If you are really keen you can pull out the boost cut pots.   The circuit will then just pass
  signal as each stage looks like an inverter.

When you mentioned the supply voltages were low a few posts back I kind of assumed the unit was working.  It is possible the extra current is because of a problem, or a faulty opamp.    Extra current can also be caused by oscillations.

[jfrabat]

Went through the whole breadboard again.  Found some errors, and corrected them (apparently, if you leave out some resistors, the pedal does not work as it should; who would have thought!).  Anyway, if there are any other errors, I have not caught them.  And I am still getting issues.

Now the signal is actually making it through, and bands 2, 3 and 4 are working as they should.  But band 1 is not cutting or boosting.  Also, if I turn the pot all the way up or down, it cuts the entire signal off very suddenly.

Here is a short video:

Now, mind you, the speaker I am using is the guitar amp, so forget about HiFi!!  LOL! 

So, in summary, everything is working as it should except Band 1's 2 issues:

• No cut/boost function
• Sound cutting at min and max (all audio signal)

Any ideas how to troubleshoot these 2 issues?

[jfrabat]

A couple of things up front.  Make sure all opamps are getting power, both + and -.   You should check the DC voltages on all opamp pins.  They should all be less than 0.1V, probably less than 20mV.

If the DC voltages are wrong you need to sort that out first.    If the DC voltages are out in an earlier band, say band 1, it will affect the DC voltages from that point on because the stages are DC coupled.

All OpAmps are gettingaround +11.75V and -11.75V.  I will check Band 1's OpAmp in all the pins shortly.

It's possible your band-4 opamp had a problem and when you moved the circuit that opamp is now on band-1 and stopping everything.

I switched the pot, but did not think about switching the OpAmp...  I will try with a new one.

Yes it is expected to get warm.  Since your circuit is pulling twice the expected current it has made things a little worse.

Doing that is fine.

Good to know!

Perhaps check the DC voltages first.  They often show-up problems.

If you have wiring issue then obviously that's going to cause problems and doing a thorough check by eye certainly helps.

When you have a largish circuit like this sometimes it is useful to get one band working at a time.    You do that by wiring the buffer to the input of each EQ stage (don't forget to disconnect the input of the EQ from the previous stage.) then wire the EQ stage output to the output jack.

Will go over Band 1.

See how you go after checking the wiring.   (I can't quite make out the wiring from the photo but when I first looked it looked like pins 1 and 3 are not going to the right place in all cases.)

You can try a few other things:
- disconnect the wipers of the boost/cut pots and see if the circuit comes to life.
  Once you disconnect the boost/cut wipers the circuit should just pass signal.
- If you are really keen you can pull out the boost cut pots.   The circuit will then just pass
  signal as each stage looks like an inverter.

When you mentioned the supply voltages were low a few posts back I kind of assumed the unit was working.  It is possible the extra current is because of a problem, or a faulty opamp.    Extra current can also be caused by oscillations.

Let me try with the new OpAmp in Band 1, and we will see

[jfrabat]

When you mentioned the supply voltages were low a few posts back I kind of assumed the unit was working.  It is possible the extra current is because of a problem, or a faulty opamp.    Extra current can also be caused by oscillations.

When you are right, you are right... 

Checked voltages of OpAmp for Band 1, and the first amplifier was reading around -10V all around (pins 1, 2 and 3).  Replaced it, and now Band 1 works.  Also, voltage is now up +13.6V for V+ and -13.5V for V-.  I will replace the resistors to bring it down to 12V again.

[Rob Strand]

Checked voltages of OpAmp for Band 1, and the first amplifier was reading around -10V all around (pins 1, 2 and 3).  Replaced it, and now Band 1 works.  Also, voltage is now up +13.6V for V+ and -13.5V for V-.  I will replace the resistors to bring it down to 12V again.

I was just about to post something then you made a new post.

Good news indeed, well done!

My guess is 150ohm and 120ohm might do it.

[jfrabat]

Checked voltages of OpAmp for Band 1, and the first amplifier was reading around -10V all around (pins 1, 2 and 3).  Replaced it, and now Band 1 works.  Also, voltage is now up +13.6V for V+ and -13.5V for V-.  I will replace the resistors to bring it down to 12V again.

I was just about to post something then you made a new post.

Good news indeed, well done!

My guess is 150ohm and 120ohm might do it.

I JUST finished with that same combo.  +12.6V and -12.5V.  All done!

THANKS TO ALL FOR THIS GREAT JOURNEY! 

Now, time to send to have the boards made!!!

[Rob Strand]

Very cool.

[jfrabat]

Just for the sake of documentation, here is the final schematic:



And this is the board layout in Eagle:



Dual-gang pots, and all jacks are off board.  Idea is to fit it in a Hammond 1590Q enclosure (in theory, it fits; will make sure once the boards arrive!).

And this is how the boards will look like:

[Rob Strand]

I just noticed the schematic has both A and B tapers for the Boost/Cut pots.   
They should be B taper (Linear).

[jfrabat]

I just noticed the schematic has both A and B tapers for the Boost/Cut pots.   
They should be B taper (Linear).

Absolutely!  And they are in the BB.  I will edit the schematic above to show it correctly.  And actually, the dual-gang ones are A also (could not find W in dual gang)

[jfrabat]

Hi, guys.  One last question about this.  I built the PCB, and have the enclosure drying.  A couple of things came out unexpectedly, but I managed to work all but one of those out.

1. My frequencies were in the opposite order (lows to the right, highs to the left), but I ended up flipping the board around to fix it (had to use a bit longer wires, but that is about it).
2. This also helped another issue; my switches were also backwards (up was down/bypass, down was off/engaged)
3. 1000uF caps were a little too tall for the enclosure, so I had to desolder them and use new ones but laying them out flat instead of sticking up.  Problem solved!
4. Even though I downloaded the template from Mouser, turns out the 3PDP switch did not fit in the holes; I had to cut the switch leads and solder legs from components used in the board.  It is not a pretty fix, but the switch is working, and it is soldered to the board!  This, of course, led to the switch being a bit higher than intended (the legs I soldered are not PERFECTLY centeres!), so I had to do a similar mod (only no cutting) on the on/off switch so that it would be at the same height (pots I could get at the right height without extensions).
5. This is the one that is throwing me off...  Everything is working, but if I boost to max the highs, and select the highest frequency, I get a high pitched tone coming out.  It only happens in the last pot, and only at max boost with the highest frequencies.  In the last Wein Bridge I used 10K instead of 12K resistors in the other 3; should I go back to 12K in the highest frequency?

Here is the board:



I will post a pic of the finalized pedal once I finish it.  Maybe even a video.