Bass Beat Detector Help

[ParkingLotLust]

Hey guys,

Ive built a "Bass Beat Extractor" and controller from here:
sound.westhost.com/project62b.htm

and it worked. But then it stopped working, and now I can't figure out how to make it work again. Ive replaced the trimmer (VR1) with another pot, and also replaced IC2, as Im not getting the signal I expect out of it.

The expected output is a waveform sitting at VCC until there is a bass beat, at which time Id see a negative pulse to trigger the 555. Right now it just sits at VCC.

Keep in mind, it was working at first, I used it for about 3 hours, and then the lights started to "stick" on. They would stay lit up for a few seconds and then go out. Eventually they just stayed on and wouldnt go off. After about five minutes they went off and we're at the present.

Any ideas would be appreciated. I have a scope and can measure/test anything suggested. Thanks!

 

[OldTechie]

What kind of circuit are you using between the "Bass Beat Extractor" and the lights? You could be trying to get too much drive out of the extractor circuit. Or, the interface between the extractor and the lights could be bad. What's the 555 circuit?

Ball's in your court now. Need more info.

Cheers,
Dave M

 

[ParkingLotLust]

The circuit Im using is straight from the next schematic on his site. Ive attached my Eagle schematic so you can see exactly what I've used. Id just like to reiterate that again, it worked fine when I first hooked it up... Its got me baffled!

Cheers Dave

 

[BrownOut]

Hey, cool project. I would definitely think the problem is in the last state, the one-shot 555 circuit. Measure the voltage on pin 2 and pin 6. They should be changing with the bass beat. As well, measure the other pins just to make sure none of the connections have opened/shorted. Make the pin 2, 6 measurements with your scope. If pins 2 doesn't toggle, then we have to start tracing back.

 

[OldTechie]

What voltage level is Vcc? If the voltage at pin 2 of IC2 isn't high enough to exceed the pin 1 voltage, the comparator (IC2) won't change state at its output. The voltage divider (R15, R16 and VR1) supplies the reference voltage for comparator IC2. The output of the voltage divider depends on the value of Vcc.
What kind of signal do you see at pin 2 of IC2? It should be a varying positive DC voltage that follows the level of the bass frequencies. What is the peak value of this envelope? You'll see it much better with an analog voltmeter ahan with a digital meter. Measure the DC voltage at IC2 pin 1. Does the peak voltage at pin 2 exceed the pin 1 voltage? If not, then you're in the right area.
Either the signal voltage from the first two IC's isn't getting amplified enough or the envelope detector isn't working. Are the diodes OK? If you have low signal voltage here, then you need to check the first two ICs.

Cheers,
Dave M

 

[BrownOut]

In your schematic, C8 is 2.2uF, in the original it is 22uF.

 

[ParkingLotLust]

Hey, cool project. I would definitely think the problem is in the last state, the one-shot 555 circuit. Measure the voltage on pin 2 and pin 6. They should be changing with the bass beat. As well, measure the other pins just to make sure none of the connections have opened/shorted. Make the pin 2, 6 measurements with your scope. If pins 2 doesn't toggle, then we have to start tracing back.
------
In your schematic, C8 is 2.2uF, in the original it is 22uF.

Thanks! Ive gotten lots of compliments from people who have seen it in action.

Here's the board itself:
**broken link removed**

And here it is driving an LED bar.

Anywho, back to troubleshooting!

Pin 2 doesnt toggle. I believe I wasnt clear enough in my first post; I said:

"The expected output is a waveform sitting at VCC until there is a bass beat, at which time Id see a negative pulse to trigger the 555. Right now it just sits at VCC."

I should have been more clearer about the fact that this expected waveform is from the output of IC2 in my schematic (sorry about the numbering being off, it came from a last-second package change). I know /TR of the 555 needs to be triggered low to fire my LEDs, and that isnt happening.

As for the value of C8, this is one of a few values I changed when I first prototyped the circuit on a breadboard. That, along with C9 and C12 gave me the specific look I was after.

What voltage level is Vcc? If the voltage at pin 2 of IC2 isn't high enough to exceed the pin 1 voltage, the comparator (IC2) won't change state at its output. The voltage divider (R15, R16 and VR1) supplies the reference voltage for comparator IC2. The output of the voltage divider depends on the value of Vcc.
What kind of signal do you see at pin 2 of IC2? It should be a varying positive DC voltage that follows the level of the bass frequencies. What is the peak value of this envelope? You'll see it much better with an analog voltmeter ahan with a digital meter. Measure the DC voltage at IC2 pin 1. Does the peak voltage at pin 2 exceed the pin 1 voltage? If not, then you're in the right area.
Either the signal voltage from the first two IC's isn't getting amplified enough or the envelope detector isn't working. Are the diodes OK? If you have low signal voltage here, then you need to check the first two ICs.

Cheers,
Dave M

VCC is ~12.2v. By pin 1 of IC2, do you mean pin 3 (the +ve input)? Pin 1 is for offset nulling on the TL051 (unless Im looking at the wrong diagram, in which case ignore me!)

I believe you are correct though, and this area is where I traced the problem to. The voltage on pin 3, the "reference" voltage if you will, varies from 8.4v to 10.1v, but the DC level of the envelope signal is no higher than 2v. If it matters, the input signal (output from the amplifier pre-out) is about 250mVpp, but it goes up to 500mVpp depending on the volume.

This tells me I dont have enough gain in the early stages (or I need to lower my reference voltage). I would like to do both - make my reference adjustment able to be much wider. But I would also like slightly more gain (or maybe selectable gain?).

Correct me if Im wrong, but the gain of a Sallen-Key is 1, so the only gain in this circuit is coming from R6/R7 (in my schematic). However, how does C4 affect the gain? I would assume it causes gain to be higher at certain frequencies (Xc = 1/[2*pi*f*C]). Whats the best way to fix my reference adjusting problem?

 

[ronv]

Seems like maybe the pot is open on pin 3. The voltage there should only be around a volt or 2.

 

[ParkingLotLust]

Ive removed the SMD pot and soldered a leaded pot in its place for now, to rule out the potentiometer itself. If I measure between pin 3 and ground, I see from 6.50kohms to 9.90kohms depending on the pot. Again, if I measure between pin 3 and ground, I see a varying voltage as the pot is adjusted (it just won't go as low as I expect it to/as low as I measure that it goes on the breadboard setup)

 

[ronv]

Maybe there is a short across R17 or the op amp is bad. Might lift pin 3 and see if the voltage goes down.

 

[ParkingLotLust]

From pin 3 to pin 6 I see 267.2k ohms. I will lift the pin and see if the voltage changes.

EDIT oh, Ive already swapped out this opamp once by the way (to rule out the opamp itself)

 

[ronv]

Hmmm. Well if its not the op amp there must be something else tied to pin 3 or 1 of the resistors in the divider cause with pin 6 high the voltage on pin 3 should be about .5 to 1.45 volts.

 

[BrownOut]

Pin 2 doesnt toggle. I believe I wasnt clear enough in my first post; I said:

"The expected output is a waveform sitting at VCC until there is a bass beat, at which time Id see a negative pulse to trigger the 555. Right now it just sits at VCC."

I should have been more clearer about the fact that this expected waveform is from the output of IC2 in my schematic (sorry about the numbering being off, it came from a last-second package change). I know /TR of the 555 needs to be triggered low to fire my LEDs, and that isnt happening.

That was my bad. I was not looking at the schematic when I made my remarks, and I thought the LED being on meant the timer was self-triggering, but that's not the case.

As for the value of C8, this is one of a few values I changed when I first prototyped the circuit on a breadboard. That, along with C9 and C12 gave me the specific look I was after.

I'll think more about that, but there is something else that seems more important.

The voltage on pin 3, the "reference" voltage if you will, varies from 8.4v to 10.1v,

That tells me there is something wrong with the voltage divider made from R15, R16 and VR1. The maximum voltage you should see is a volt or so. Check the color code of those resistors very carefully. Still doesn't explain why it worked before though, so some mystery remains. But I know what you're reporting here doesn't seem right.

Correct me if Im wrong, but the gain of a Sallen-Key is 1, so the only gain in this circuit is coming from R6/R7 (in my schematic). However, how does C4 affect the gain? I would assume it causes gain to be higher at certain frequencies (Xc = 1/[2*pi*f*C]). Whats the best way to fix my reference adjusting problem?

C4 should not affect the gain. It's main purpose is to help cancel out DC offset.

 

[ParkingLotLust]

Success! I replaced R15 and R16, and Im back to working! I wonder what went wrong...

So now Ive got an adjustable reference in the range I want. Whats the easiest way to bump up the high-end of the voltage by half a volt or so?

EDIT actually, how hard would a more broad range, say .5 - 5v instead of .5 - 1.5v, be? As it is, if the volume is turned up too loud, I cant get the reference adjusted high enough

EDIT 2 wait, I can just increase the pot to 50k to get a more broad range. durrr!

 

[BrownOut]

Which voltage? If you mean signal voltage, you can try to replace R7 with a 500ohm pot, or a combination of a fixed resistor, about 100 ohms + a pot that gives 500 ohms total.

PS: Forget what I just wrote, I did't understand your question. I'll look at what you're trying to do.

 

[BrownOut]

You can make VR1=100k. That would give you both more range and a higer top voltage. It will be a little harder to make fine adjustments, however.

 

[tufty88]

Hey,

I'm trying to make this circuit- however I want to use a microphone as an input, instead of using a direct line signal. Does anyone know how I would edit the schematic to accomodate a microphone input?

Thanks,
Jonathan

 

[audioguru]

I'm trying to make this circuit- however I want to use a microphone as an input, instead of using a direct line signal. Does anyone know how I would edit the schematic to accomodate a microphone input?

You should learn that a microphone has a very low output level and it needs a preamp circuit to increase its level up to a normal line level. Line level is about 220mV and a mic has a level of about 5mV so the preamp needs a signal voltage gain of about 44 times.

 

[Benik33]

Hi.
I want also build this circuit and you are the only one, who I found that he build it

So my question is, do you have some recommendation for this schematic? (What change or what should I be careful?)
Also do you have please the PCB and schematic in Eagle?

Thank you

 

[Benik33]

OK, I made a schematic and board in Eagle for the beat extractor.
I'm not expert in it, but better then nothing
[MODNOTE]Link deleted[/MODNOTE]