Wien bridge oscillator math - diyAudio
 Wien bridge oscillator math
 User Name Stay logged in? Password
 Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read Search

 Equipment & Tools From test equipment to hand tools

 Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
 20th February 2013, 10:59 PM #3 diyAudio Member   Join Date: May 2011 Location: Charlotte NC Thanks for the insightful answers! That's an enormous help and lots to ponder on. And I was planning on doing the sine to square to triangle wave path, though the triangle wave is a bit wonky at his point. I have purchased some op amps with a higher slew rate which I'm hoping will help. I can't seem to edit my first post so I'll repost the missing link: Wein Bridge Oscillator - Wikipedia, the free encyclopedia Thanks again! Patrick
 23rd February 2013, 04:22 PM #4 diyAudio Member   Join Date: Jan 2012 Location: Staines, UK My guess of a possible answer to question D) is that the cheap capacitors were ceramic and the capacitance changed with voltage which helped control the feedback. The large value multilayer type can have a large reduction in capacitance with voltage. I have found Oscillators difficult to stabilize because the amplitude change with time is an exponential of the Oscillation frequency x gain error.
 23rd February 2013, 11:59 PM #5 diyAudio Member     Join Date: Sep 2010 Location: Grapeview, WA Hi Patrick -- the Wikipedia file shows the simplest form of a Wien bridge oscillator, and this form can be found in a Linear Technology App Note on oscillators and bridges, although I can't at the moment remember the number. The reactive legs of the bridge (assuming equal Cs and equal Rs) results in an AC attenuation of 3, while the resistor + lamp leg has to have approximately the same attenuation -- this means for a lamp, which is not intended to run at full operating voltage, but rather at around 10% of rated voltage, that has an operating resistnace in circuit of say 150 ohms, the resistor will have to be 2x that value or 300 ohms. If you're using an opamp like the LM49710 or LT1037, then I suggest using the 1869 type lamp, available from Mouser. The reactive feedback path blocks the DC from the output, while the resistive feedback path sets the DC output level of the amp. If you want the oscillator output to be zero volts DC, then use an opamp with split power supplies. To use the single-ended version and get maximum output swing, return the bulb to a low impedance voltage divider at half the positive supply which is bypassed at the bulb by a large electrolytic cap for low AC impedance to ground. Use a pot in series with the feedback resistor to adjust for lowest THD and best stability. See my webpage on lamps for oscillators -- www.moorepage/Lamps.html __________________ ................... Dick Moore
diyAudio Member

Join Date: May 2011
Location: Charlotte NC
Quote:
 Originally Posted by PChi I have found Oscillators difficult to stabilize because the amplitude change with time is an exponential of the Oscillation frequency x gain error.
Thank you for your input. I definitely concur! I am trying to calibrate the oscillator to handle a range of about 200Hz to 20 kHz and it's not easy task. Probably a strong argument for using multiple switched capacitors, however...

Quote:
 Originally Posted by richiem If you're using an opamp like the LM49710 or LT1037, then I suggest using the 1869 type lamp, available from Mouser. The reactive feedback path blocks the DC from the output, while the resistive feedback path sets the DC output level of the amp. If you want the oscillator output to be zero volts DC, then use an opamp with split power supplies.
I am using a Radio Shack lamp rated at 5V, 25mA. I can see the "bounce," it's definitely doing something, so I'm not sure if it's part of the problem. I also have a 12V version which I haven't tried yet. My supply is currently 12V, single-ended.

I have rewired the circuit using the op-amp version instead of the 4QD version. My first observation is... it doesn't really work! I think I failed to consider the amount of current the op amp has to drive due to the lamp. My first try was with a LM833 and that does oscillate and the wave looks very clean, but it's only oscillating at about 2V p-p; that wouldn't really be a big problem except that I feel I'm overloading it. I also tried an NE5532 and a TLC2272; neither of those oscillated at all with the lamp. If I removed the lamp and put in place a 10K resistor and a pot for the feedback resistor it oscillated just fine. That makes me think it's a current issue.

So I don't know at the moment if I'll buy some higher current op amps, perhaps those Dick suggested, or if I'll go back to the 4QD circuit. Just thinking out loud here (so to speak) but I guess two other avenues would be using one of my life-time-supply of Silonex optocouplers in place of the lamp or some hybrid solution with some kind of transistor buffer between the output and the resistive leg of the feedback loop. Don't know if the latter's possible but I seem to remember seeing something like that.

As far as the single supply goes, I'm mimicking the circuit in figure 15-10 in this link:
http://www.ti.com/lit/an/slod006b/slod006b.pdf

I'm using a second op amp to generate a 2V reference voltage in place of "Vref." This is supposed to bring my output's midpoint to 6V.

Quote:
 Originally Posted by richiem To use the single-ended version and get maximum output swing, return the bulb to a low impedance voltage divider at half the positive supply which is bypassed at the bulb by a large electrolytic cap for low AC impedance to ground.
I like this idea better because it's based on half the supply, but I can't get it to simulate correctly. I attached the circuit I _think_ you were describing but I may have gotten something wrong

Thanks for all the input guys!
Attached Images
 Wien_OpAmp_1sply.gif (12.8 KB, 180 views)

diyAudio Member

Join Date: Feb 2012
Location: Havana, Cuba
I built this one. It only needs one pot to change frequency. Don't know the amount of distortion, and have to say that amplitude varies with frequency.
Attached Images
 Generador de Audio.jpg (51.1 KB, 164 views)

diyAudio Member

Join Date: May 2011
Location: Charlotte NC
Thanks for posting the circuit; that looks great! I'm kind of set on using the lamp, but this might work well too.

I attached my current circuit for the oscillator core. I added a transistor buffer past the op amp and it's helping a whole lot with the calibration. It's a lot more stable within a pretty large frequency range. It doesn't do well past about 12k so I'm thinking that has to do with current delivery from the voltage reference op amp as the resistors get small -- I've seen some voltage sag.

I'm really hoping to get this working without switching any caps, though, as I want to control the frequency with optocouplers driven by a microcontroller (see attached). I'm going to set up an AGC the same way, too, but that will be _after_ the selected waveform (square, sine or triangle) is switched into the next stage of the circuit.
Attached Images
 WienCore_bufferedOut.GIF (5.1 KB, 155 views) micro.jpg (151.9 KB, 153 views)

 6th March 2013, 07:53 PM #9 diyAudio Member   Join Date: Sep 2004 Location: Montreal a google search produced this nice read and circuit... http://www.janascard.cz/PDF/An%20ult...0-140%20dB.pdf /paba
 7th March 2013, 02:02 AM #10 diyAudio Member   Join Date: May 2011 Location: Charlotte NC You read my mind. I was just thinking about that! I have tweaked and calibrated my circuit to where I am now getting my full target range of 100Hz-20kHz without cap-switching. This is using two optocouplers in the wien bridge sharing a current source and attached to a pot. But the last couple days I've been thinking about how I might go about swapping the lamp out for a third optocoupler. Especially, I'm thinking if I could reduce the output level to a smaller value, that would be better for overall distortion caused by the optocouplers; then re-amplify with an op amp. With a lamp, it has some pretty fixed properties and you have to work around that but optocouplers seem much more controllable as far as how sensitive you want them to be, what kind of time constant you want them to have, etc., since one piece is completely isolated from the other. I've only found one other similar circuit to this so this is a great find that I will definitely check out further.

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are Off Pingbacks are Off Refbacks are Off Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Mark A. Gulbrandsen Digital Line Level 19 2nd November 2014 03:35 AM akis Equipment & Tools 10 18th October 2012 10:15 AM mod_evil Equipment & Tools 42 5th July 2008 06:17 PM pjpoes Analogue Source 15 27th December 2006 06:10 PM

 New To Site? Need Help?

All times are GMT. The time now is 08:39 PM.