Help with OpAmp Clean Boost

[Box_Stuffer]

I am trying to make a clean boost using a TL072 with a charge pump. The idea is to make 1 side a distortion and 1 side the clean boost and then blend them. I am using the charge pump in place of a voltage divider, but when I set up the gain stage it is not working as it should. When I ground the negative feedback loop through a resistor it stops working. If I have only a resistor between pin 6 and 7 it works, but I don't know how to calculate the amount of gain it is creating.

Can I use a single reference voltage for two separate inputs? I just thought the charge pump would be less messy but it seems to cause other issues.

[FiveseveN]

Please always post a schematic of exactly what you're doing, charge pump and all. And ideally clear photos of the build.

[Box_Stuffer]

This is the charge pump.





The signal goes into this splitter and comes out as separate signals.




Then I was trying to make a boost based on something like this.




When I set it up like this it does not work. If I have only R1, I get a signal.

[GGBB]

Please post a schematic of how you connected everything together and pics of the build.

[GibsonGM]

Read this, report back :https://www.diystompboxes.com/smfforum/index.php?topic=29816.0

And - take each part as separate from the others, and testable...meaning, does the charge pump circuit output the correct voltage, before connecting to anything else? 

[Box_Stuffer]

The 1 Spot power supply puts out 9.5v and the voltage inverter IC puts out -8.7v.

[Rob Strand]

When I ground the negative feedback loop through a resistor it stops working. If I have only a resistor between pin 6 and 7 it works, but I don't know how to calculate the amount of gain it is creating.

You need a cap in series with resistor, checkout the TS-9 circuit for example.  Cap chosen to taste.

The Doo doo labs Sprinkle Drive has a structure close to what you want,
https://www.oocities.org/fabricadetimbres/imagens/doodoolabsdrive.jpg
(FWIW, I think the final circuit ended up with feedback diodes, two in each direction.)

Can I use a single reference voltage for two separate inputs? I just thought the charge pump would be less messy but it seems to cause other issues. 

Often you don't need the series cap in that case.  If you have a lot of DC gain it can amplify the opamp's DC offset and cause all sort of trouble , perhaps even enough DC offset at the output to cause the diodes to clip!!

[m_charles]

you gotta be able to at least know the basics of schem reading (if you don't already) to get much help here. Those stripboard diagrams aren't much help. You can learn how to read basic schems in a day if you dig into it.

[antonis]

If I have only a resistor between pin 6 and 7 it works, but I don't know how to calculate the amount of gain it is creating.

Easily..

Just put "infinite" in Gain formula denominator..
(unity gain buffer..)

P.S.
As Rob said above, you DO need both a resistor and a capacitor to GND..
The resistor sets the gain (together with feedback one) and the capacitor ensures DC gain roll-off to unity..
(capacitor blocks DC which results into unity gain..)
Without that cap in series with gain resistor, your 800mV DC off-set [9.5 - (-8.7)] is multiplied by stage's gain on op-amp output..
(you can easily realize that for a gain greater than 10X your circuit is permanently hitting power supply rail..)

[Box_Stuffer]

I set up a gain stage exactly as in the bottom picture of the Gain Stage diagram. I used a 100K for R1 and a 1k for R2 to make a gain of 100, but with a linear pot it does not turn down all the way and when I turn it up past half way the sound gets really harsh and screechy.

[Box_Stuffer]

When I ground the negative feedback loop through a resistor it stops working. If I have only a resistor between pin 6 and 7 it works, but I don't know how to calculate the amount of gain it is creating.

You need a cap in series with resistor, checkout the TS-9 circuit for example.   Cap chosen to taste.

The Doodoo labs Sprinkle Drive has a structure close to what you want,
https://www.oocities.org/fabricadetimbres/imagens/doodoolabsdrive.jpg
(FWIW, I think the final circuit ended up with feedback diodes, two in each direction.)

Can I use a single reference voltage for two separate inputs? I just thought the charge pump would be less messy but it seems to cause other issues. 

Often you don't need the series cap in that case.  If you have a lot of DC gain it can amplify the opamp's DC offset and cause all sort of trouble , perhaps even enough DC offset at the output to cause the diodes to clip!!

So if I abandon the charge pump altogether and I want to keep the signals into each side of the opamp totally separate, do I need to make different reference voltage points for each side? Thats all I really need to know and I can probably go from there.

[GGBB]

I set up a gain stage exactly as in the bottom picture of the Gain Stage diagram. I used a 100K for R1 and a 1k for R2 to make a gain of 100, but with a linear pot it does not turn down all the way and when I turn it up past half way the sound gets really harsh and screechy.

That sounds normal. The op-amp configuration is non-inverting, so gain is 1 + R1/R2 - meaning at minimum unity gain. "Harsh and screechy" is probably just op-amp clipping - too much gain for the voltage supply which is what limits the output signal. Suppose a guitar input signal is 250mV p-p (it will depend on the pickups and many other factors) - at 100 gain that tries to become 25V p-p but your supply only has ~18V. Note also that most op-amps clip before the power supply rail limit, losing roughly a couple of volts of headroom.

[Rob Strand]

So if I abandon the charge pump altogether and I want to keep the signals into each side of the opamp totally separate, do I need to make different reference voltage points for each side? Thats all I really need to know and I can probably go from there.

Like most pedals you can have a single Vref provided you have a big enough cap bypassing it.

[antonis]

What Rob said..

[Mark Hammer]

Missing from this entire discussion is just how MUCH boost you require or are aiming for.  Certainly, if the objective is to goose the input signal by 40x, then use of a charge pump to improve the headroom is recommended.  But if the objective is more modest - say 5x, to get a level-hike for solos - then +9v may well be quite sufficient, unless you have super-hot pickups, or are sticking the boost after something that had already boosted the level.

[Box_Stuffer]

Not really sure what the end goal is exactly because I am learning as I go. I did say that I want to blend the Overdrive and Clean signals, so it has to be comparable to the output of the OD circuit. I am just using the Wampler basic OD that looks like this :




I am using 1N60P germanium diodes and they soak up a lot of the signal, but not as much as the 1N34s so I like them better.

I was probably making WAY more gain that necessary for the boost. It is hard to find many examples of opamp boosts. Most are transistor or JFET. All I could find was the BBE Boosta Grande and very basic gain stage diagrams.

[antonis]

"Boost" is usually a stage (or more than a single one) that amplifies the signal amplitude up to a desirable level without employing distortion..

For ordinary signal levels and 9V supply, a gain of 4X (12dB) should be considered OK..

P.S.
For your particular configuration, I'd propose R1 = 33k, R2 = 10k and Rp = 10k..
(max - min boost -> 12.7 - 8.5 dB)

[antonis]

It is hard to find many examples of opamp boosts.

Let's get back..

[Box_Stuffer]

It is hard to find many examples of opamp boosts.

Let's get back..

Thanks for the help - as always.  I'll play around with some of these suggestions and see what happens.