Frequency response does not match project

Thread Starter

AmiguelS

Joined Jun 25, 2014
20
Hi!
I few weeks ago i projected this circuit, which you guys helped me debug (it had a offset problem). Today I built it and while testing i noticed it was attenuating higher frequencies of the audio spectre (I audio tested it as I do not have an available oscilloscope at the moment).

Can you please help me out? I don't seem to be able to spot the problem.
Multisim says it should have a flat magnitude from +-10Hz up.

Thanks!




EDIT: The "source" circuit is the one bellow.

 

MikeML

Joined Oct 2, 2009
5,444
I assume you perceive that the high frequency response is lacking?
How can you tell if you are not using a scope?
 

Thread Starter

AmiguelS

Joined Jun 25, 2014
20
I assume you perceive that the high frequency response is lacking?
How can you tell if you are not using a scope?
It's behaving like it had a low-pass filter cutting frequencies of interest.
I can tell because this is to be used in a guitar pedal board, and I tested it that way and the treble drop is quite perceivable.

Lack of sanity.
Thanks for caring about my sanity but I believe it to be just fine.

test the two 10 micro farad caps, if one has dried out, it will have a very small capacitance.
I will test them asap, but wjat do you mean dried out? Would using polarized caps instead for the non polar indicated make a difference?

Thank you all for the responses.
 

MikeML

Joined Oct 2, 2009
5,444
If C1 or C3 was dried out (lower than normal capacitance), that would effect the low frequency response. If you are lacking highs, then there is something else going on.

The opamps have a gain-BW product of 55MHz, so you cant blame them. The 1K gain setting resistors R10 and R19 are too low; 10K would be better.

Rather than trying to get all the gain in one stage (U1), it would be better to distribute the gain in both U1 and U2. If you need a total gain of 100, better to get a gain of 10 followed by a gain of 10...
 

Thread Starter

AmiguelS

Joined Jun 25, 2014
20
If C1 or C3 was dried out (lower than normal capacitance), that would effect the low frequency response. If you are lacking highs, then there is something else going on.

The opamps have a gain-BW product of 55MHz, so you cant blame them. The 1K gain setting resistors R10 and R19 are too low; 10K would be better.

Rather than trying to get all the gain in one stage (U1), it would be better to distribute the gain in both U1 and U2. If you need a total gain of 100, better to get a gain of 10 followed by a gain of 10...
I see. I guess I'l try that. What is bothering me is that up until now the simulation has always been spot on (it even confirmed some mistakes i made previously while assembling the circuit) and it shows everything should be fine.

What about the polar/non polar matter?

Off-topic: Is non polar and bipolar the same?
 

MikeML

Joined Oct 2, 2009
5,444
If using polarized (electrolytic) capacitors, C1 needs to have the + end on the right, and C3 needs to have the + end on the left.
 

THE_RB

Joined Feb 11, 2008
5,438
Lack of drive at the output?

If a circuit has weak drive (high output impedance) then the result will be a loss of HF when amplified. The amplifier will load down the circuit at HF.
 

Thread Starter

AmiguelS

Joined Jun 25, 2014
20
Lack of drive at the output?

If a circuit has weak drive (high output impedance) then the result will be a loss of HF when amplified. The amplifier will load down the circuit at HF.
How can I get more drive then? Shouldn't the OpAmp have a low output impedance already?
 

THE_RB

Joined Feb 11, 2008
5,438
Test it. Make a 1v p/p output sinewave, then load it down with different load resistors to see how much the output drops.

When you find a resistor that loads the 1v signal down to be 0.5v that will roughly match the output impedance.

(edit) Sorry just realised you don't have a 'scope.

What kind of amp are you using?
 

Thread Starter

AmiguelS

Joined Jun 25, 2014
20
Test it. Make a 1v p/p output sinewave, then load it down with different load resistors to see how much the output drops.

When you find a resistor that loads the 1v signal down to be 0.5v that will roughly match the output impedance.

(edit) Sorry just realised you don't have a 'scope.

What kind of amp are you using?
I'm using a hybrid amp, valve pre amp, SS power (Orange Micro Terror).

I'm have a pseudo oscilloscope using an Arduino, its not good for measurements only for wave shape comparing. I can use a multimeter though...
 
Last edited:

kubeek

Joined Sep 20, 2005
5,794
What is your source signal? Is it coming straight from the guitar pickups? Try adding a buffer between C1 and S1.
 

Alec_t

Joined Sep 17, 2013
14,280
I can tell because this is to be used in a guitar pedal board, and I tested it that way and the treble drop is quite perceivable.
Perhaps whatever you plugged it into for the test could be attenuating the high frequencies?
 
Top