Wired up a phase shift oscillator but I dont think im getting a sinusoidal wave

[Qaiphyx]

I hooked it up to a speaker but the speaker just makes some noise, flexes up and stays up, then shutters a little and repeats, no sound wave it seems. It should be at 900hz I have a dual + - 9 volt battery supply. Pin 1 is empty as it is just offset. Pin 2 has incoming from pin 6 and is bridged over the OP amp to pin 6 also. Pin 3 is grounded. Pin 4 has - input. Pin 5 is empty. Pin 6 is the output going to res cap res cap res cap res then back to pin 2, 3 resistors are grounded. Pin 7 has the + input. Pin 8 is empty. I wired the speaker after the end resistor before it goes to pin 2 from pin 6, so it should have a full phase shift. Any ideas?

 

[Qaiphyx]

Would the type of capacitor make any difference? The caps are from a few weeks ago and I think they might be electrolytic if that makes any difference?

 

[Hero999]

How about posting the schematic and we might have some idea of what you are talking about?

 

[Qaiphyx]

Ive been using this.

http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/oscphas.html

 

[Qaiphyx]

Just waiting for the mods to approve the post. I use the schematic from hyperphysics of the phase shift oscillator.

here

http : // hyperphysics . phy-astr . gsu.edu / hbase / electronic / oscphas . html

to speed things up.

 

[Hero999]

Why did you add the spaces?
Here's the link without the spaces.
**broken link removed**
http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/oscphas.html
You do know that this circuit can't drive an 8Ω speaker, you need a buffer amplifier like the LM386.

 

[Qaiphyx]

Why did you add the spaces?
Here's the link without the spaces.
**broken link removed**
http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/oscphas.html
You do know that this circuit can't drive an 8Ω speaker, you need a buffer amplifier like the LM386.

The impediance of my speaker is 4Ω. LM386?

The forum wont allow me to post links with out approval since im a new member.

 

[Qaiphyx]

The impediance of my speaker is 4Ω, will it push that or is that worse? lol. LM386?

The forum wont allow me to post links with out approval since im a new member.

 

[Qaiphyx]

I can try to buffer it with the other op amp that I have but do you think this is definitely the problem? Eventially I will be buffering it.

 

[Hero999]

Adding another op-amp won't help since the minimum impedance most op-amps can drive is 2k, some can drive 600R but even that's 150 times too high.

Making the speaker 4Ω makes things worse, an LM386 can't drive that, you need to either use an 8Ω or a higher power amplifier IC like the TDA2003.

 

[Qaiphyx]

Adding another op-amp won't help since the minimum impedance most op-amps can drive is 2k, some can drive 600R but even that's 150 times too high.

Making the speaker 4Ω makes things worse, an LM386 can't drive that, you need to either use an 8Ω or a higher power amplifier IC like the TDA2003.

Hmm I guess it must be working then. I guess ill just have to get another speaker then for frequency testing.

Thanks alot for you help!

Also for this circuit, is it possible to make a + - power supply like I have with my 2 9 volt batteries with just one battery?

 

[Hero999]

You can run this circuit off a single supply with some simple modifications.

 

[Roff]

You can run this circuit off a single supply with some simple modifications.

That circuit does not take into account the biasing of the output due to the connection to ground of the rightmost R. Neither takes into account the AC loading of the phase shift network by R1.
Below is a circuit that deals with both these problems, and also has a soft limiting circuit to avoid the waveform clipping that will otherwise occur. Simulation shows 3rd harmonic distortion of ≈-43dB.

 

[Qaiphyx]

That circuit does not take into account the biasing of the output due to the connection to ground of the rightmost R. Neither takes into account the AC loading of the phase shift network by R1.
Below is a circuit that deals with both these problems, and also has a soft limiting circuit to avoid the waveform clipping that will otherwise occur. Simulation shows 3rd harmonic distortion of ≈-43dB.

Will it still push a frequency with out these changes? These are good changes, I will have to use them. My first goal is to get it working first. Then Ill perfect it

 

[Roff]

Will it still push a frequency with out these changes? These are good changes, I will have to use them. My first goal is to get it working first. Then Ill perfect it

If you want to use Hero999's circuit, I would modify it as below.

 

[Qaiphyx]

If you want to use Hero999's circuit, I would modify it as below.

Alright.

I just realized that im using all the same resistors for everything. but in the schematics its showing r1 greater than r2 whats the explination of this?

as you can tell im fairly new to electronic circuits.

 

[Roff]

Alright.

I just realized that im using all the same resistors for everything. but in the schematics its showing r1 greater than r2 whats the explination of this?

as you can tell im fairly new to electronic circuits.

You have it backwards. The schematic shows R2 greater than 29 times R1. I added the note in red, which means R1 should be much greater than R. I would try R=3.3k, R1=33k, R2=1Meg. The first circuit I posted has fewer constraints.

 

[Qaiphyx]

You have it backwards. The schematic shows R2 greater than 29 times R1. I added the note in red, which means R1 should be much greater than R. I would try R=3.3k, R1=33k, R2=1Meg. The first circuit I posted has fewer constraints.

Ok thanks. Yah I had a bunch of 1 k resistors and that was it ill have to run to radioshack tomorrow lol. Would this mistake prevent oscillation?

Also, what should I use for a frequency around 3400hz?

 

[Roff]

Ok thanks. Yah I had a bunch of 1 k resistors and that was it ill have to run to radioshack tomorrow lol. Would this mistake prevent oscillation?

Also, what should I use for a frequency around 3400hz?

You can use the 1k resistors for R. Use 10k for R1, and 300k for R2. To get around 3400Hz, see the original link:

f=1/(2*pi*R*C*√6)
R*C=1/(2*pi*f*√6)
R*C=1.911*10^-5
If R=1k=1000,
C=19.11nF
You can use either 18nF or 20nF to get close to 3400Hz.

 

[Roff]

You should change R2 to 330k. In a simulation, it would not run with 300k, but with 330k, and 18nF caps, it gave me 3.5kHz with very low harmonic distortion (-50dB). You might want to use a 270k or 300k resistor in series with a 50k or 100k pot, and crank the pot value down to where it just breaks into oscillation. This will give you the least distortion.