Trouble with active HPF (high pass filter)

[fezder]

Heyy!
I'm designing light organ, it's done only but filter section left to do
I do bandpass filter for detecting different amplitudes of audio frequencies, i have no trouble so far doing LPF section of bandpass filter, but with same cutoff HPF section somehow causes opamp output to have it's lower/minus side of sine wave to be cut off, even if input voltage is correctly biased for this opamp, LM358, it should be adequate for audio speed, but is there are other fault?
simple circuit i'm using:
edit, sorry about topic typo, filter cutoff is 30hz i'm using, if that matters, just for testing

**broken link removed**

 

[Nigel Goodwin]

What value components are you using?, and why choose such a very low frequency?.

It's also about the poorest possible filter you could use.

 

[jjw]

If you have single supply voltage, you need a virtual ground for the circuit.
For better filters search active second order filters, eg. Sallen-key etc.

 

[audioguru]

The highpass filter circuit you have is so simple that 30Hz is reduced only a little, 15Hz is reduced only a little more, 7.5Hz is reduced only a little more and 3.75Hz is reduced only a little more. But it still passes earthquake rumbling frequencies so what is it for?
Your schematic shows an opamp that is missing an important power supply. I suspect the supply is positive only so of course since its input is grounded then negative parts of waveforms are cutoff. If the power supply is positive and negative voltages then the output will be fine, except the lousy old LM358 is noisy (hiss) and produces crossover distortion.

 

[Mikebits]

Hello Fezder, Give this online filter design tool a try. It is made by Analog Devices so it should be pretty good.
https://www.analog.com/designtools/en/filterwizard/#/type

 

[audioguru]

Why do you say your simple filter is active? It is not active, it is a simple passive RC filter followed by an opamp follower.
An active filter uses some of the filtering parts as part of the negative feedback.

 

[KeepItSimpleStupid]

High pass? the gain will be reduced at the corner frequency, so would that not be low pass? (passes the lower frequencies below Fc)

or am I confused?

 

[audioguru]

The reactance (AC resistance) of the series capacitance increases at low frequencies that reduces their levels, therefore it passes high frequencies.

 

[crutschow]

................
An active filter uses some of the filtering parts as part of the negative feedback.

Or in general, uses some form of feedback to shape the response, such as the positive feedback used in a Sallen-Key active filter.

 

[fezder]

What value components are you using?

capasitor is 10n, and ressitor is 5.6k

and why choose such a very low frequency?

well i just took 30hz for measuring, i most likely use say 60hz or so in final after i get this sorted out, 30hz is indeed very low

It's also about the poorest possible filter you could use.

you mean due the fact it's first order? yeah, i thought it would be enough for start before i add more poles

Your schematic shows an opamp that is missing an important power supply

i'm sorry, i forgot to tell that most important, it's one-side, +12v and ground,

Why do you say your simple filter is active?

oh my! i thought filter comes active when there is something that uses power rails in any manner....

 

[fezder]

here's shots to perhaps explain situation better
so only difference in circuit is that i swapped resistor and capacitor in between
LPF, works as intented, no clip at any frequency i tested

HPF, with clip, with any frequency

 

[jjw]

Can you show the schematics, specially biasing of the opamps + input.
I am guessing that the input voltage has a positive offset and in lpf case the + input gets bias from this offset and in hpf the + input is connected to 0V.

 

[fezder]

there's nothing more to show in schematics,only biasing i've made is in input, function generator is offset so there is no negative signal at all going in opamp. so yes, input voltage has positive offset

 

[Nigel Goodwin]

i'm sorry, i forgot to tell that most important, it's one-side, +12v and ground,

You need to generate a split supply for it then, there are examples in the opamp 'sticky' at the top of this forum.

https://www.electro-tech-online.com/threads/basic-opamp-circuits.35258/

 

[fezder]

hmm, i must use/make split supply even if lm358 is single supply? so there's then +-12v and ground, like putting batteries in series i get that far, but since this opamp won't like negative voltage, what does that negative 12v do? bit confusing for me.....

 

[fezder]

well i'll be, made that virtual ground and it did the trick, cutoff gone. Now, i just dont understand, why and how?

 

[fezder]

But, i tested it with low-pass, and it does that clipping there with virtual ground being used?

 

[jjw]

there's nothing more to show in schematics,only biasing i've made is in input, function generator is offset so there is no negative signal at all going in opamp. so yes, input voltage has positive offset

But with hpf there is no positive offset to + input of the opamp because the input capacitor cuts it and you have input voltage that is between
+Vinpeak -Vinpeak

 

[fezder]

Ah, of course! btw, i used virtual ground as input ground, like where i ground either resistor or capacitor, i suppose i did that right?

 

[atferrari]

Hola fezder
Besides doing an eventual reading to grasp the basics, you could download from Microchip their filter lab which works fine.

Good to have it in your PC.