Is there anyone who tried this circuit? I would like to add volume control on output and maybe some gain control on first opamp stage...
(https://i.postimg.cc/qtzCDXrK/Active-LPF-Final-TB-Version-1.jpg) (https://postimg.cc/qtzCDXrK)
Nope, haven't tried it, but your additions seem easy enough to do. The first stage is a non-inverting op-amp, so you'd just add a couple of resistors, something like this:
(https://electricdruid.net/wp-content/uploads/2019/01/InputBufferSchematic.jpg)
For the volume control, you could replace the current 220K to ground at the output with a pot instead, and then take the output from the pot's wiper.
HTH
I was thinking about same thing as you, i try to make pcb layout for it...what is difference between connecting R8 on your example to ground and vref?
For single supply circuits, you really want to reference everything to half supply. Which then goes to ground. You can either do this or shunt it to ground through a capacitor. But you can't just shunt it to ground, you'd have actual DC current flowing through the feedback essentially.
Quote from: Domin on March 03, 2024, 02:44:58 PMI was thinking about same thing as you, i try to make pcb layout for it...what is difference between connecting R8 on your example to ground and vref?
+1 what fryingpan said.
The difference is you need to add a capacitor if you're going to connect it to ground. The input bias resistor (R1/220K on your schematic, R6/2M2 on mine) sets the DC bias point for that op-amp at Vref, so if you then tie the feedback network to ground, you've got a big DC difference, and the amplifier will think that's an input and try to amplify it. However, if you put a cap in series, it blocks the DC and it'll work.
It would be good practice to add *two* caps to that stage. There'll be one from R8 down to ground which also acts as a highpass filter, so can be sized up to help cut mains hum and subsonic thumps. The other one would go in parallel with R7 and sets a lowpass cutoff to limit the high frequencies (well, at least roll-off the gain for high frequencies to only x1). That should be sized for the highest audio frequency of interest. For Bass, that could probably be pretty low - maybe 5-8KHz, an octave lower than I'd usually go for guitar?
With moderate gains, these parts aren't critical, but as gain goes up, it becomes more and more important to make sure that amp stages are only amplifying stuff you *want* amplified, and not loads of other rubbish that you don't (subsonics, ultrasonics, radio frequencies, etc etc).
Quote from: fryingpan on March 03, 2024, 06:17:34 PMFor single supply circuits, you really want to reference everything to half supply. Which then goes to ground. You can either do this or shunt it to ground through a capacitor. But you can't just shunt it to ground, you'd have actual DC current flowing through the feedback essentially.
Then why almost every opamp circuit in guitar pedals is connected to ground? Like DOD 250 etc? I think that connecting to vref i found in all Bajaman's designs but not in common circuits...
Quote from: ElectricDruid on March 03, 2024, 06:26:39 PMQuote from: Domin on March 03, 2024, 02:44:58 PMI was thinking about same thing as you, i try to make pcb layout for it...what is difference between connecting R8 on your example to ground and vref?
+1 what fryingpan said.
The difference is you need to add a capacitor if you're going to connect it to ground. The input bias resistor (R1/220K on your schematic, R6/2M2 on mine) sets the DC bias point for that op-amp at Vref, so if you then tie the feedback network to ground, you've got a big DC difference, and the amplifier will think that's an input and try to amplify it. However, if you put a cap in series, it blocks the DC and it'll work.
It would be good practice to add *two* caps to that stage. There'll be one from R8 down to ground which also acts as a highpass filter, so can be sized up to help cut mains hum and subsonic thumps. The other one would go in parallel with R7 and sets a lowpass cutoff to limit the high frequencies (well, at least roll-off the gain for high frequencies to only x1). That should be sized for the highest audio frequency of interest. For Bass, that could probably be pretty low - maybe 5-8KHz, an octave lower than I'd usually go for guitar?
With moderate gains, these parts aren't critical, but as gain goes up, it becomes more and more important to make sure that amp stages are only amplifying stuff you *want* amplified, and not loads of other rubbish that you don't (subsonics, ultrasonics, radio frequencies, etc etc).
What do you think is better way to do it? Connecting to vref or to ground with 2 caps creating a filter?
Quote from: Domin on March 04, 2024, 12:54:41 AMwhy almost every opamp circuit in guitar pedals is connected to ground? Like DOD 250 etc? I think that connecting to vref i found in all Bajaman's designs but not in common circuits...
(https://i.postimg.cc/4HV5KSLt/DOD250-no-ground.gif) (https://postimg.cc/4HV5KSLt)
Quote from: Domin on March 04, 2024, 12:57:07 AMQuote from: ElectricDruid on March 03, 2024, 06:26:39 PMQuote from: Domin on March 03, 2024, 02:44:58 PMI was thinking about same thing as you, i try to make pcb layout for it...what is difference between connecting R8 on your example to ground and vref?
+1 what fryingpan said.
The difference is you need to add a capacitor if you're going to connect it to ground. The input bias resistor (R1/220K on your schematic, R6/2M2 on mine) sets the DC bias point for that op-amp at Vref, so if you then tie the feedback network to ground, you've got a big DC difference, and the amplifier will think that's an input and try to amplify it. However, if you put a cap in series, it blocks the DC and it'll work.
It would be good practice to add *two* caps to that stage. There'll be one from R8 down to ground which also acts as a highpass filter, so can be sized up to help cut mains hum and subsonic thumps. The other one would go in parallel with R7 and sets a lowpass cutoff to limit the high frequencies (well, at least roll-off the gain for high frequencies to only x1). That should be sized for the highest audio frequency of interest. For Bass, that could probably be pretty low - maybe 5-8KHz, an octave lower than I'd usually go for guitar?
With moderate gains, these parts aren't critical, but as gain goes up, it becomes more and more important to make sure that amp stages are only amplifying stuff you *want* amplified, and not loads of other rubbish that you don't (subsonics, ultrasonics, radio frequencies, etc etc).
What do you think is better way to do it? Connecting to vref or to ground with 2 caps creating a filter?
You are still creating a filter if you reference everything to Vref, because you are going to have a shunt capacitor in parallel with the voltage divider.
A funny thing happened to one of my designs: if I just set the volume to zero (after a gain stage) according to simulations I'm actually still getting the signal, 30dB down, and heavily filtered. That's because the pot is referenced to Vref, and the voltage divider + capacitor are too high an impedance for the signal to be properly shunted to ground, so some of the signal bleeds back into the circuit essentially. (It's not a real issue, I chose to reference the pot to Vref so that I could use one fewer resistor, not good practice but it should work for an overdrive pedal).
Best practice would be:
- use Vref only to set half supply voltage, it's not a true "ground" for the signal;
- shunt signal, whenever signal is actually meant to be flowing through the node, to ground through a capacitor to maintain the DC "bias" point.
Note that bias setting resistors will still have signal flowing through them, so you do need to shunt the signal through ground (through a capacitor next to the voltage divider).
Any designer using VRef for purposes other than its existence (BIAS...!!) is illusively relied upon absolutely no DC offset between various stages.. :icon_wink:
e.g.
In an ideal world DOD250 design, C3 & C4 could be omitted and clipping diodes lower end could be connected to Vref..
Quote from: antonis on March 04, 2024, 05:26:48 AMAny designer using VRef for purposes other than its existence (BIAS...!!) is illusively relied upon absolutely no DC offset between various stages.. :icon_wink:
e.g.
In an ideal world DOD250 design, C3 & C4 could be omitted and clipping diodes lower end could be connected to Vref..
DC offset is often inconsequential. The opamp I'm employing states 0.3mV typical, 3mV maximum. Could be lower, but it's not too bad.
Quote from: Domin on March 04, 2024, 12:57:07 AMWhat do you think is better way to do it? Connecting to vref or to ground with 2 caps creating a filter?
I think the better way to do it is to take it to ground with the cap. Like PRR said, you should only use the Vref bias supply for bias. Generally, the *less* stuff you have connected to Vref, the better.
Secondly, limiting the gain of the op-amp to only the bandwidth of interest is good practice, and the caps help with that.
Quote from: fryingpan on March 04, 2024, 05:56:36 AMDC offset is often inconsequential. The opamp I'm employing states 0.3mV typical, 3mV maximum. Could be lower, but it's not too bad.
I'm talking about voltage drop across bias resistor..
500nA (max) through 470k resistor -> 235mV..
If you were to be very careful about offset voltage and current, you would have to fill the board with (very large) capacitors. If Rod Elliot is fine with leaving them out, who am I (a hobbyist) to object?
Somehow i made this layout, it doesnt have values yet, still thinking about better design...i would like to have something more profesionnal
(https://i.postimg.cc/FfMYDSp3/Screenshot-20240304-165358.jpg) (https://postimg.cc/FfMYDSp3)
Quote from: Domin on March 04, 2024, 10:46:32 AMSomehow i made this layout, it doesnt have values yet, still thinking about better design...i would like to have something more profesionnal
(https://i.postimg.cc/FfMYDSp3/Screenshot-20240304-165358.jpg) (https://postimg.cc/FfMYDSp3)
After fixing errors and adding values
(https://i.postimg.cc/FYC6CMz5/Screenshot-20240304-171115.jpg) (https://postimg.cc/FYC6CMz5)
I havent build It, but googleing the schem info shows thatthe designer made a video too in post 83 https://www.talkbass.com/threads/a-different-kind-of-low-pass-filter.1146653/page-5
Sounds good to me in his own
Quote from: sergiomr706 on March 04, 2024, 02:17:37 PMI havent build It, but googleing the schem info shows thatthe designer made a video too in post 83 https://www.talkbass.com/threads/a-different-kind-of-low-pass-filter.1146653/page-5
Sounds good to me in his own
Thats where i got schematic from
Is there anyone who can make better layout with onboard pots? Or some better placement for wiring them?
Quote from: Domin on March 05, 2024, 01:29:21 AMIs there anyone who can make better layout with onboard pots? Or some better placement for wiring them?
We *could*, but then you'd never get any practice, and you'd never improve.;)
Start by placing the pots where you want them on the board, then put the IC on, then do the power, and only then route the other stuff.
Quote from: ElectricDruid on March 05, 2024, 07:10:46 AMQuote from: Domin on March 05, 2024, 01:29:21 AMIs there anyone who can make better layout with onboard pots? Or some better placement for wiring them?
We *could*, but then you'd never get any practice, and you'd never improve.;)
Start by placing the pots where you want them on the board, then put the IC on, then do the power, and only then route the other stuff.
I tried but it would have a lot of jumpers :D But ill give it a try, other were worse than what i posted here
Quote from: ElectricDruid on March 05, 2024, 07:10:46 AMQuote from: Domin on March 05, 2024, 01:29:21 AMIs there anyone who can make better layout with onboard pots? Or some better placement for wiring them?
We *could*, but then you'd never get any practice, and you'd never improve.;)
Start by placing the pots where you want them on the board, then put the IC on, then do the power, and only then route the other stuff.
I tried but it had a lot of jumpers :D Best is what i posted here
If you're trying to do it as a single-sided board for etching yourself, then it's going to need a few jumpers, so don't worry about it. Trying to completely avoid them usually means twisting things around in a such a way (and routing long tracks all over the board) that it makes the layout worse, not better. Get the component layout nice and tidy and don't cramp things up.
If you're doing a board for a prototype PCB place to make for you, do it double-sided. Something like that would be a nice little thing to start with and you should be able to do a decent job. These days, there's no real advantage to doing a single-sided board. Everything below four layers is so basic for PCB fab places these days, they regard it all as stone-age tech!
Quote from: ElectricDruid on March 05, 2024, 12:18:20 PMIf you're trying to do it as a single-sided board for etching yourself, then it's going to need a few jumpers, so don't worry about it. Trying to completely avoid them usually means twisting things around in a such a way (and routing long tracks all over the board) that it makes the layout worse, not better. Get the component layout nice and tidy and don't cramp things up.
If you're doing a board for a prototype PCB place to make for you, do it double-sided. Something like that would be a nice little thing to start with and you should be able to do a decent job. These days, there's no real advantage to doing a single-sided board. Everything below four layers is so basic for PCB fab places these days, they regard it all as stone-age tech!
I did a lot of boards, prototyped or etched but this one gives me a headache :D
Quote from: ElectricDruid on March 05, 2024, 12:18:20 PMIf you're trying to do it as a single-sided board for etching yourself, then it's going to need a few jumpers, so don't worry about it. Trying to completely avoid them usually means twisting things around in a such a way (and routing long tracks all over the board) that it makes the layout worse, not better. Get the component layout nice and tidy and don't cramp things up.
If you're doing a board for a prototype PCB place to make for you, do it double-sided. Something like that would be a nice little thing to start with and you should be able to do a decent job. These days, there's no real advantage to doing a single-sided board. Everything below four layers is so basic for PCB fab places these days, they regard it all as stone-age tech!
(https://i.postimg.cc/n9FTjKSf/Screenshot-20240305-193015.png) (https://postimg.cc/n9FTjKSf)
New one, probably best placement i could do, opinion?
Quote from: Domin on March 05, 2024, 01:31:27 PMNew one, probably best placement i could do, opinion?
Nothing like.
Everything is over to the left. All the resistors are squashed up into 0.3" spacing. 0.4" is a lot more comfortable.
Move the IC a couple of squares (0.2") over to the right and re-do it. It'll have more space and be more balanced.
The current layout has lotsof stuff on the left and almos nothing except a few capacitors on the right. It needs to be more-or-less equal between the two sides.
HTH
Quote from: ElectricDruid on March 05, 2024, 06:42:34 PMQuote from: Domin on March 05, 2024, 01:31:27 PMNew one, probably best placement i could do, opinion?
Nothing like.
Everything is over to the left. All the resistors are squashed up into 0.3" spacing. 0.4" is a lot more comfortable.
Move the IC a couple of squares (0.2") over to the right and re-do it. It'll have more space and be more balanced.
The current layout has lotsof stuff on the left and almos nothing except a few capacitors on the right. It needs to be more-or-less equal between the two sides.
HTH
(https://i.postimg.cc/Hr53YBF3/Screenshot-20240306-174818.jpg) (https://postimg.cc/Hr53YBF3)
Something like that?
Quote from: Domin on March 06, 2024, 11:49:47 AM(https://i.postimg.cc/Hr53YBF3/Screenshot-20240306-174818.jpg) (https://postimg.cc/Hr53YBF3)
Something like that?
Yeah, that's getting better, don't you think?
Quote from: ElectricDruid on March 06, 2024, 01:24:06 PMQuote from: Domin on March 06, 2024, 11:49:47 AM(https://i.postimg.cc/Hr53YBF3/Screenshot-20240306-174818.jpg) (https://postimg.cc/Hr53YBF3)
Something like that?
Yeah, that's getting better, don't you think?
I think its too big now :D
Quote from: Domin on March 06, 2024, 01:30:51 PMQuote from: ElectricDruid on March 06, 2024, 01:24:06 PMQuote from: Domin on March 06, 2024, 11:49:47 AM(https://i.postimg.cc/Hr53YBF3/Screenshot-20240306-174818.jpg) (https://postimg.cc/Hr53YBF3)
Something like that?
Yeah, that's getting better, don't you think?
I think its too big now :D
So squash it up a bit. There's still plenty of space.
The caps along the top could go up one. The 47Ks and the electros at the bottom could go up one. The five components on the left should *definitely* go up one, so they're directly below the 6K8s. Once you've done all that, I think you could move the IC and almost everything up one.
Making it less wide is harder, but if you move that 470K one to the right, the blue trace wouldn't have to go all the way around the outside. And the track to the 6K8 could come under the pot like the other one does. You can't get it narrower than that without moving the pots closer together.
You could probably move the big electros at the bottom up and to the right a bit so they don't do down below the 47Ks. Then you could eliminate more rows off the bottom.
No trace proximity issues..??
(https://i.imgur.com/6ldb7HX.jpg)
Quote from: antonis on March 07, 2024, 05:07:31 AMNo trace proximity issues..??
(https://i.imgur.com/6ldb7HX.jpg)
I can change it, no problem
Are they that big of a problem? Over half a millimetre should be enough after all at audio frequencies. And parasitic effects can and should be dealt with through filtering, I think.
Quote from: fryingpan on March 08, 2024, 06:38:21 AMAre they that big of a problem?
Not for me.. :icon_wink:
Made this preamp with some mods but im not much satisfied with results, i hoped for something better and switch somehow doesnt work
(https://i.postimg.cc/Pv1tvrGv/IMG-20240314-095704-298.jpg) (https://postimg.cc/Pv1tvrGv)
(https://i.postimg.cc/4YfJngWn/IMG-20240314-095713-238.jpg) (https://postimg.cc/4YfJngWn)