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.

Sallen Keys filter slowly dies

Status
Not open for further replies.
I have been working on a circuit that lights up LEDs when the bass hits.

Here are the steps in my circuit.

1) Class A amplifier

2) Op Amp

3) Sallen Keys low pass filter

4) Inverting Amplifier

5)FET

I thought I had finally got it working, but after putting it into a box, I noticed that after 1-2 minutes, the cutoff frequency becomes so low that the LED no longer light.

I have been trying to figure this out for hours, and the only thing I have come across is that if I connect one of the junctions in the filter to ground with my finger, the circuit will work again for another 1-2minutes. Any help would be greatly appreciated. -Alex
 
if a picture is worth a thousand words, a schematic is worth pages of descriptive text posts
 
We can't tell anything without a schematic, of course, but my wild-ass guess is that there's a capacitor somewhere that's getting slowly charged and changing the behavior of the filter (or of another part of the circuit). We'll see how good my guess is.

BTW, it's "Sallen-Key". Like so many things electronical, it's the names of the people who developed it.
 
Last edited:
Another wild guess. If you are using CMOS linear op amps and don't have proper biasing an input to op amp so it is floating that gradually puts op amp output to supply rail. This could also happen if you are using DC coupling and have a temp induced bias shift in early stage.

You can check the bias voltage on each stage output.
 
Last edited:
Your first two op amp stages + inputs are floating. Put a resistor to ground on each op amp + side and 10 uF cap connection.

You actually don't need coupling caps between the op amps, just the from the bipolar input stage and first op amp. With gains of unity for first two stages there will not be any DC offset amplification.

I also assume you are running pos and neg supplies to op amps.

You are also going to need a resistor to ground on the MOSFET input otherwise it will be ON all the time.

The LED in opto coupler is acting as a rectifier on output of last stage. Check its reverse breakdown voltage. Should be greater then negative supply voltage.
 
Last edited:
Explanation of the circuit:

Class A amplifier to amplify the voltage

OP amp to amplify the current

Sallen-Key filter to cutoff certain frequency

Inverting amplifier to change the negative voltage to positive

Optocoupler to vary the brightness depending on how loud the sound is

FET to drive the large amount of LED's I have.



I designed this at school but had limited time so getting it working was the most important goal.
The signal was a near perfect replication of the line-level audio with a gain of 10, until the lowpass filter.
The lowpass filter seemed to really mess up the signal.
Using a scope, the output with no signal was almost the applied voltage. With a low frequency, the negative peaks would dip down to about zero, with no positive alternation. We couldn't figure out why this was happening, so we just fixed the problem and put this signal into an inverting amplifier.

Also, I have a large capacitor across the bus. When I disconnect power, short the busses, and reconnect power, it still will not work. I have to disconnect power and wait 1-2 minutes to reconnect power for it to work again.
I am using a CMOS LM324 OP amp, so the idea that a voltage on one of the OP amp's input sounds like the problem, as I don't know of any capacitor on the circuit that would be slowly charging without discharging, except maybe one of the coupling capacitors? Anyway, thanks for your help and I will appreciate any more help you are willing to give.

Also, for doing projects like this and maybe some radio broadcasting circuits, how important is a oscilliscope? I would really like to have one, but can't really jusitfy spending $300 for a hobby.
 
Your first two stages + inputs are floating. Put a resistor to ground on each op amp + side and 10 uF cap connection.

I also assume you are running pos and neg supplies to op amps.

Yes I am supplying the OP amps with the correct voltages.
When you say put a resistor to ground on the op amp, are you referring to the input? And what about the sides and a 10uf cap?
 
Yes I am supplying the OP amps with the correct voltages.
When you say put a resistor to ground on the op amp, are you referring to the input? And what about the sides and a 10uf cap?

Put a 50k resistor to ground on first op amp + input and change the 10 uF coupling cap to 0.1 uF. Eliminate the 10 uF between first two op amps, just direct connect. Eliminate 0.1 uF between second and third op amp, just direct connect.

Put a 22k to ground on input of MOSFET. You may also want to put a 1nF cap across the 22k to ground on MOSFET input. This will act as a filter cap for the rectified output, otherwise the LED may be too dim without it.

You don't really need the first op amp buffer stage.
 
Last edited:
Thanks

Thanks, I missed the 10k to ground I have on the gate of the FET. I'm going to add the 50k on the input of the filter. Will post results
 
Last edited:
Thanks for all your advice, it sounds like it will work.

However, after drawing this schematic, I tried to reconnect this circuit to make sure it still (partially) worked before I modified it.

After hooking everything up, nothing happened. At school I would just trace the signal with a scope to see where the problem is. Seeing as I can no longer do this, how can I trace the problem? Or do I just need to shell out for a scope?
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top