Bad output wave-form from my filter??

[stl_jones@yahoo.com]

This is my first project on my own and is really annoying me.

My main goal is to turn an LED on and off with low frequencies with 12V and 0V rails.

I send the amplified RCA signal to the filter. There is a sine-wave on the input of the op amp. However, the output does is at 12V DC with no signal. With a low frequency, the output dips below 12V to about 0V. However, this alternation is only about 1/4 the time that the output is at 12V (20% duty cycle?).

I can't understand why this is happening and am 1000% sure I have it hooked up correctly.

I am using an LM324 op amp, 20k resistors on the filter, .01uF on the filter, and 10k feedback resistors on the filter.

Please help as this is driving me insane.

**broken link removed**

 

[carbonzit]

How are you powering the 324: single supply? split supply? With a single supply, you'll need a voltage divider (1:1) to eliminate the DC offset to the op amp.

 

[stl_jones@yahoo.com]

supply

yes, I am using a single supply.

Where should I put the voltage divider? The output I'm assuming

 

[audioguru]

Your Sallen and Key lowpass filter must be fed from a very low impedance like the output of an opamp.
An opamp with a dual-polarity supply has its input biased at half the total supply voltage which is 0V. Then its output can swing equally up and down.

I added an opamp input buffer that has a very low output impedance and it is biased at half the supply voltage. It is DC-coupled to the input of the filter so the filter opamp is also biased at half the supply voltage.
I added a capacitor to ground at the feedback of the filter so it does not amplify the DC bias voltage.

 

[stl_jones@yahoo.com]

Ok thanks.

I'm trying just to get the first transistor to amplify the signal. However, I've had 0 success using this schematic. Can someone help me to get the correct component values to get this amplifier working. Also, where do you go to learn these things? I've had no luck trying to find a good tutorial online. thanks

**broken link removed**

 

[carbonzit]

If you really want to learn "these things", I'd say forget online tutorials. What you need is a good electronics textbook. There are several threads here discussing such books. My current favorite is Malvino's Electronic Principles. There are many other good ones.

You also need to be able to do the math. No true understanding of electronics without an equal understanding of the math.

I know: I have a broad overall knowledge of electronics, but since I never properly studied it, I lack understanding of many fundamentals. I'm trying to educate myself as well.

 

[audioguru]

The output impedance of a transistor amplifier is too high to feed the lowpass filter. It also has an input impedance which might be too low for your signal source to drive.
Use an opamp instead like I showed. The input opamp can have as much gain as you need.

The old LM324 quad opamp is noisy (hiss), has crossover distortion and has poor response above only 2kHz. Use a much better quad opamp.

 

[carbonzit]

The old LM324 quad opamp is noisy (hiss), has crossover distortion and has poor response above only 2kHz. Use a much better quad opamp.

Once again, Audio "guru" totally misses the point with an unhelpful reply.

After all, the O.P. did say:

My main goal is to turn an LED on and off with low frequencies with 12V and 0V rails.

Notice the "low frequencies"? "turn an LED on and off"? Who cares about crossover distortion, noise or poor response above 2kHz in this case?

 

[stl_jones@yahoo.com]

It's alright, he's just trying to help. I got it to work using a 2NNNNA transistor instead of a c945.

Could you recommend a good electronics book? We had a great one at school but I forgot the name of it.

Off-topic, but I am almost done, I just need a way to turn on my High-powered LED's. I need to use a transistor as a switch that is normally ON. I assume a JFET would be best for this application. Any other suggestions are welcomed

I have a whole lot of audio equipment laying around, are there any JFET's in audio equipment? If not, what everyday-electrical device might have one? (I don't want to wait 3 days to order one online)

Thanks!

 

[audioguru]

Once again, Audio "guru" totally misses the point with an unhelpful reply.

Notice the "low frequencies"? "turn an LED on and off"? Who cares about crossover distortion, noise or poor response above 2kHz in this case?

You are correct, I missed the point because there was no reason given to blink an LED with a sine-wave at a frequency (up to about 1600Hz which will not be filtered much) that is much higher than we can see (frequencies above only about 40Hz do not appear to blink).

 

[Jaguarjoe]

Wikipedia explains transistor biasing quite well. Look for it under the same name.

The Art of Electronics is a good book. eBay has paperback editions pretty cheap.
You can get Malvino's book on Amazon for about $7
Both are older copies but the theory hasn't changed very much in the last few years.

 

[wkrug]

Additional you should wire the unused Op Amp's of the LM324.

Connect the Output to the - Pin and give 1/2 Supply Voltage to the + input.
The Voltage divider you can make with 2 10k Resistors and an 100nF cap.

When you left open one or more Op Amps it can be oscillating, and the whole Chip wouldn't work.

 

[carbonzit]

Additional you should wire the unused Op Amp's of the LM324.

Connect the Output to the - Pin and give 1/2 Supply Voltage to the + input.
The Voltage divider you can make with 2 10k Resistors and an 100nF cap.

I would think that tying both inputs of any unused op amps would be more than sufficient. A little paranoid to think you need to tie the output down.

 

[wkrug]

I've meanth the inverting input of the according Op Amp,
and not the negatve supply voltage of course.
And that's marked in the Data Sheet with "-".

 

[Brevor]

wkrug's method is considered the proper way to do it.

 

[MrAl]

Hi,

Because this filter has a gain of 2 at DC, we can use a 4:1 voltage divider to bias the very input and use a capacitor to couple from the preceding section to this new section. That should be good enough. The upper resistor of the voltage divider can be something like 300k and the lower resistor 100k. The coupling cap could probably be 0.1uf or so. There will be a small low frequency cut, but it wont affect the audio frequencies.
I dont see any simpler way to bias this thing, although there are probably other circuits that can be used too.

 

[stl_jones@yahoo.com]

Ok thanks a lot. I've been having a little trouble making a transistor switch that is normally on. I was going to use a 2n2222a transistor as an inverter, then send the output to another transistor, but there is way too much lag. It takes at least 1/2 seconds for the inverter's output to return high after the signal is killed. Any idea's or should I just try to find a JFET?

I am using a 10k on the base of the inverter, grounded the emitter, and 1k on the collector. Thanks

 

[audioguru]

A transistor used as an inverter switches in a few tens of nano-seconds, not hundreds of milli-seconds.
Since your inverter switches extremely slowly then you must be doing something very wrong.

 

[carbonzit]

A transistor used as an inverter switches in a few tens of nano-seconds, not hundreds of milli-seconds.
Since your inverter switches extremely slowly then you must be doing something very wrong.

Well, duh.

So to the O.P. (stl_jones): This really cannot be due to your transistor, so I'm guessing there's a capacitor involved somewhere that's either charging or (more likely) discharging during that long 1/2-second delay. Any capacitors in your circuit? is it the circuit in this post? can you post a schematic with component values?

 

[ronv]

My guess is that there is more to the circuit than you have shown us in your schematic. Having said that, in any case you need to go back and build the circuit in post #4 if you want to get a signal out.