Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Notch Filter Woes...

Status
Not open for further replies.

nfway

New Member
Hello all,
I'm new to speaker design, so please pardon the ignorance.

I have built a small guitar amplifer that feeds a 8 ohm speaker, and am now fine tuning the output to my needs. Can anyone tell me how I might be able to filter out 800 Hz completely? I've plugged a equalizer pedal into the amp and layed with it until I narrowed it down to this frequency.

The amp sounds great with this frequency filtered out, but not so hot with it in.

I've tried the Twin T design on the following page:
**broken link removed**

I used value of 1.8K for R and .1uF for C which, according to the calculations should have been a good starting point, but the output was too low, and very distorted. I'm not sure if this has anything to do with what is going on, but the amplifier uses BTL configuration, therefor the outputs are floating with respect to ground.

Thanks in advance for any and all helpful suggestions.
 
My understanding of simple filters is that they attenuate to some degree but do not block completely. You might need several stages.
 
The twin-t notch by itself has very low Q, which means the notch is very broad. You can improve it dramatically with bootstrapping (feedback). See this app note. I wouldn't use such large resistor values (and small caps). I would use R=20k and C=10nF. Make the pot 10k. You don't have to use the op amps shown. I would use something like LM324 (a quad op amp), and add another voltage follower (output tied back to inverting input, as in the other two in the schematic) in front of the twin-t. You still have one more op amp, which you need to tie as a voltage follower and ground the non-inverting input to prevent it from possibly upsetting the internal biasing of the other three. You will have to tell us what power supply(s) you have available, and what the DC level of your input is. Below are simulations of the circuit with the parts I recommended. The three traces are for the pot at the bottom of its range (yellow, no feedback), at 80% rotation, and at 99% rotation. The circuit doesn't work well at 100% rotation because of the op amps' low bandwidth, so you might want to add 100 ohms in series with the top end of the potentiometer. You might also want to add a resistor in series with the grounded leg of the pot to limit the range.

Important: Don't try to put this between your amp and your speakers. It won't drive speakers. Put it in front of your amp.

The lower waveform is a frequency zoom of the upper one.
 

Attachments

  • twin-t_zoom.gif
    twin-t_zoom.gif
    15 KB · Views: 700
  • twin-t.gif
    twin-t.gif
    13.6 KB · Views: 672
Thanks so much for the quick responses.

Is there any way I can do this without using the opamps? I was trying to accomplish this with just an RC circuit.

I am using a 12V, 1.3aH lead-acid battery as the power source which I pull out and charge when needed.
 
AFAIK, the twin-t is the only passive RC notch circuit. LC circuits can have higher Q, but they require large inductors for low frequencies.
 
Does this mean I can build this circuit without adding the opamps, as I was doing before, or will it not work that way?

Also, is the pot in this circuit used as a volume control, or should it be additional to the volume pot?

Thanks again....
N
 
I'm not sure I understand the question. I'm assuming the 800Hz is coming from the input signal (guitar). There is nothing in my amp circuit that affects/changes the incoming signal in any way... that I know of. The input is fed through the amp circuit, and directly to the speaker.
 
nfway said:
I'm not sure I understand the question. I'm assuming the 800Hz is coming from the input signal (guitar). There is nothing in my amp circuit that affects/changes the incoming signal in any way... that I know of. The input is fed through the amp circuit, and directly to the speaker.

They should be no reason to have to filter out a specific frequency, this isn't something which guitar amps (or any other amps) generally do. Try your guitar on a known working amp, and see how it sounds - I would imagine it will sound fine.

I would suspect you have a serious design (or construction) flaw in your amplifier, or (less likely) your speaker is resonating at 800Hz. Perhaps you should post the circuit so we can see what it's doing?.
 
I think you are correct in suspecting that the speaker is resonating at 800Hz. The circuit is the "typical application" located on the chips datasheet (see attached). I just need to fine tune the speaker to the enclosure, and remove the unwanted frequency.
 

Attachments

  • circuit_730.jpg
    circuit_730.jpg
    32.5 KB · Views: 657
nfway said:
I think you are correct in suspecting that the speaker is resonating at 800Hz. The circuit is the "typical application" located on the chips datasheet (see attached). I just need to fine tune the speaker to the enclosure, and remove the unwanted frequency.

Well that looks pretty straight forward, what about the preamp?.

BTW, do you have anpther speaker you could try, or even a pair of headphones (remembering to feed them through an attenuator).
 
I'm not using a pre-amp due to the fact that my guitar (Ovation) has a built-in preamp circuit. I'm not sure what this looks like as I have not taken it apart.

I've tried several other speakers which vary in freq. response, however the overdriven 800Hz sound is prevelant in all.
 
If that schematic is an accurate representation of your amplifier, I think I see your problem. Why is there a 100uF capacitor connected from the output of amplifier 1 to ground at pin 6?
 
Ron H said:
If that schematic is an accurate representation of your amplifier, I think I see your problem. Why is there a 100uF capacitor connected from the output of amplifier 1 to ground at pin 6?

I would suspect that the diagram (which is simply a manufacturers suggested use) doesn't show all the components in the chip. I would imagine that the capacitor on pin 6 is actually decoupling the bottom of the negative feedback resistors to chassis.
 
nfway said:
I'm not using a pre-amp due to the fact that my guitar (Ovation) has a built-in preamp circuit. I'm not sure what this looks like as I have not taken it apart.

I've tried several other speakers which vary in freq. response, however the overdriven 800Hz sound is prevelant in all.

You still really need a preamp, guitars aren't intended to feed directly into power amps - the preamp in the guitar is basically to enable the use of longer cables without any loss, but more practically as an excuse to charge more for the guitar!. It's unbelievable what they can charge for a few cheap components!.

The diagram doesn't give any suggestion as to the input impedance of the power amp, and we don't know the output impedance of the preamp either, both of these could cause different effects.
 
I played with several different resistor/capacitor sets, and finally came up with a good combination using the twin-t notch filter as suggested early on. The output sounds much smoother now, and I am not getting any distortion.

Thanks to all for the suggestions and help... you guys are great!

Also, I'm not really sure what the purpose of the 100uF cap off of pin 6 is. As I said, I was just going by the "typical application" recommendation of the manufacturer.

Thanks again,
N
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top