5 string fretless showcase: Habanero V

[gyancey]

I don't understand how it's a Vref of 4.5v if it's being seen through a 1M resistor, but I'll push the 'I believe' button for now

The input impedance is very high (10^13 ohms according to the spec sheet) so ideally there is almost no current flowing through the 1Meg resistor, thus the voltage at the input of the opamp is 4.5V but in your case it won't because you don't have a blocking capacitor between the pickup output and the opamp input. Without that the pickup's DC resistance functions as a resistor to GND and the opamp input can't be biased to 4.5V.

 

[adam.beauchene]

The input impedance is very high (10^13 ohms according to the spec sheet) so ideally there is almost no current flowing through the 1Meg resistor, thus the voltage at the input of the opamp is 4.5V but in your case it won't because you don't have a blocking capacitor between the pickup output and the opamp input. Without that the pickup's DC resistance functions as a resistor to GND and the opamp input can't be biased to 4.5V.

Ah, got it, forgot to add that in the schematic TL posted. Fixed:

Thanks for catching that. I'm feeling a compulsion to, once I have a working preamp, to get some boards ripped and make them available at cost to other people here in the LC. There are a few guys doing that in the headphone amp world, why not here? I'm not very experienced at circuit design, but I have been through the process of taking a proven schematic, transcribing it into Eagle, uploading it to a board house, then sourcing components and poulating the board myself. I might start a separate thread with that in mind, to try and solicit some input from the electronically gifted around here. Thanks so far to musiclogic, Taylor Livingston, and gyancey!

 

[gyancey]

Almost - move R10 to connect directly to IC1A pin 3. Add a larger cap in series between IC1B pin 7 and the volume pot. Leave the feedback resistor R6 as is. Once you DC bias the first stage to Vref you'll have a DC offset (most likely nearly Vref) at the output that you want to DC block.

 

[pilotjones]

[tangent]

Greg

In case no one's said it yet, it's nice to see you back on TB.

[/tangent]

 

[adam.beauchene]

Almost - move R10 to connect directly to IC1A pin 3. Add a larger cap in series between IC1B pin 7 and the volume pot. Leave the feedback resistor R6 as is. Once you DC bias the first stage to Vref you'll have a DC offset (most likely nearly Vref) at the output that you want to DC block.

I realized my error in placing R10 between C7 and the pickups as soon as I posted that. It's fixed, and an output cap added. See anything else? Thanks a lot for the help!

 

[Taylor Livingston]

The output cap should generally be bigger than that. 1u to 10u. The cap on the input of an opamp is generally smaller than that on the output, due to the high input impedance and low driving impedance.

You can get 1u film caps for very cheap, so that's a good way to go. If you use an electrolytic, the "hi fi" thing to do is to use an oversized cap to limit the distortion that electrolytics can introduce, which is why we normally see 10uf caps on the outputs of opamps. In reality I doubt you'd hear a difference betwen 1 uf and 10uf electros, but it's probably a 1 cent difference, so what the heck.

 

[adam.beauchene]

The output cap should generally be bigger than that. 1u to 10u. The cap on the input of an opamp is generally smaller than that on the output, due to the high input impedance and low driving impedance.

You can get 1u film caps for very cheap, so that's a good way to go. If you use an electrolytic, the "hi fi" thing to do is to use an oversized cap to limit the distortion that electrolytics can introduce, which is why we normally see 10uf caps on the outputs of opamps. In reality I doubt you'd hear a difference betwen 1 uf and 10uf electros, but it's probably a 1 cent difference, so what the heck.

...So will too small of an output cap limit the output, because the smaller value cap can't hold a very large charge? This is fascinating stuff. I've got a ton of the green 'chiclet' looking film caps, I'll use one of those. Fixed:

 

[T-34]

Too small value could limit low frequency response in the case you plug the bass into low-impedance input (for example, direct to mixing desk witout DI; some effects like fuzzface,...)

By the way, with bigger output cap you can significantly lover the value of volume pot too, down to 10K-25 K, providing better driving capability and better rf interference protection at intermediate volume pot positions.

 

[T-34]

And another remark: you have DC flowing through your "bass" pot, that's wrong because of two facts: it can become scratchy and, more important, the DC tension on "5" input of opamp will vary with "low" pot position, throwing its bias to hell.

Better add another decoupling cap after IC1

 

[adam.beauchene]

And another remark: you have DC flowing through your "bass" pot, that's wrong because of two facts: it can become scratchy and, more important, the DC tension on "5" input of opamp will vary with "low" pot position, throwing its bias to hell.

Better add another decoupling cap after IC1

Thanks T- I wondered if DC through the tone stack mattered. So DC through tone stack is bad because it means DC through pots- I'll put that in my notes, thanks! Fixed:

It seems that there are quite a bit of caps that need to be added to deal with the DC offset from using a Vref. I know that in hi-fi, removing caps from the signal path is a design goal, because of even the best cap's tendency to alter frequency response. I also know that in musical instrument circuits, the distortions that components provide can be desirable, helping to shape the 'signature sound' of so many guitars, basses, amps and preamps. But say I wanted to reduce the number of signal path caps, just to see what it sounds like. Rather than using a Vref of 4.5v, what do you guys think about a + and -4.5v rail to power pins 4 and 8, so that the opamp swings around 0v on its own? It seems they're designed to be used that way anyway. I'm going to build this one based on all of the suggestions I've received (at this point I'm calling this a community effort), but I'll work on another schematic with power power conditioning components and less signal path components. It will be fun to compare and contrast... besides, I want a table saw and a HVLP sprayer before I start on anymore basses, but I always need a project.

 

[T-34]

R5 should still go to the "1" of IC1A

 

[adam.beauchene]

R5 should still go to the "1" of IC1A

Oops, killed my gain settings, didn't I... Fixed.

 

[T-34]

And I am bothered by your R4 and R7 - shouldn't they go to VRef instead of the ground? I cannot quite make my mind about these ...

 

[Berserkr]

Is it me, or have the last 10 or so posts been in a different language?

 

[T-34]

It's you, we understand ourself perfectly

 

[theundecidedone]

Is it me, or have the last 10 or so posts been in a different language?

definitely technical gibberish. nobody understands it, really

 

[adam.beauchene]

Is it me, or have the last 10 or so posts been in a different language?

It is a different language- One that I'm trying to learn. So far I can order a drink and find the bathroom, but I still can't order off the menu or understand directions

 

[adam.beauchene]

And I am bothered by your R4 and R7 - shouldn't they go to VRef instead of the ground? I cannot quite make my mind about these ...

R4 was part of the baxandall tone stack. [DEL]Shouldn't R7 be OK connected to ground, since the signal is swinging around Vref out of the first stage?[/DEL] Wait, I think you're right- since C9 makes the signal swing around 0v, the second stage will clip the negative side of the signal if it isn't Vref'fed.

 

[adam.beauchene]

...or should pin 5 get a blocking cap and a Vref?

 

[Musiclogic]

Adam, this thread is fantastic, not only for the build, but the extent you are going to explaining your process with the pre. Absolutely outstanding!!!