Post by thread:[EE]:unwanted hi frequency osc with opamp design ?

Hi all, I have built a volume control circuit using opamps and a digital pot. The opamps I am using are NE5532N. The input buffer and output driver amps are configured as inverting amplifiers. This is so I can add gain later if need. My problem is that as the circuit reaches unity i.e. digital pot is at 0. The input amp has a hi frequency osc. This is not audible nor does it noticabley affect the audio but should not be there. I have read somewhere that I should have some capacitance somewhere to stop this but can not find the reference :-(( Also what value input decoupling capacitor do you recomend Thanks a bundle Matt -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

> The input amp has a hi frequency osc. I don't know any details of your circuit, so this is only a general answer. Some amps are unstable without compensation, especially a low gains. Try adding a "compenstation capacitor". Some amps have explicit pins for this, otherwise put it directly from output to the negative input. Of course this assumes your circuit doesn't drive the negative input with a low impedance at high frequencies. Start with the smallest capacitor you can find and work up slowly. Usually only a few 10s of pF will do it. Too much compenstation will cut into the desired bandwidth. Of course this assumes you have good clean and well bypassed power supplies at the amp. All else is irrelevant until this is taken care of. ******************************************************************** Olin Lathrop, embedded systems consultant in Littleton Massachusetts (978) 742-9014, .....olinKILLspam@spam@embedinc.com, http://www.embedinc.com -- http://www.piclist.com hint: To leave the PICList piclist-unsubscribe-requestKILLspammitvma.mit.edu

Matthew wrote: >Hi all, > >I have built a volume control circuit using opamps and a digital pot. >The opamps I am using are NE5532N. The input buffer and output driver amps >are configured as inverting amplifiers. This is so I can add gain later if >need. My problem is that as the circuit reaches unity i.e. digital pot is at >0. The input amp has a hi frequency osc. This is not audible nor does it >noticabley affect the audio but should not be there. I have read somewhere >that I should have some capacitance somewhere to stop this but can not find >the reference :-(( > For a normal inverting opamp situation, you can prevent oscillation and also filter hi-freq noise by simply adding a small cap between output and neg-input nodes. You normally have a feedback [gain-setting] resistor [Rf] between those nodes. You select the cap value based upon the value of Rf and the BW [F3db] you desire for the ckt, using the following formula: Cf = 1/(2*pi * F3db * Rf) The cap will both filter out noise higher in freq than F3db, and also prevent hi-freq oscillations - in "most" situations. However, if you vary Rf to change gain, then this analysis is more complicated, because the BW of the ckt will vary as Rf varies. Also, if the gain-setting element is a digital pot in the feedback pathway, rather than resistor Rf, then its characteristics will affect the analysis too. - dan michaels http://www.oricomtech.com ====================== -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam.....mitvma.mit.edu

In addition to the good suggestions others have made here's something I've seen often. If the load you're driving has too much capacitance you can get oscillation on the output of most op-amps. With a LM324 type it takes only a few tens of pF to make the output oscillate. Some op-amps can tolerate a lot more load capacitance. To decouple the op-amp from the capacitive load simply add a small (10 to 100 ohm) resistor in series with the output. If the series resistance causes too much drop in the output you can move the feedback takeoff from the output pin to the other side of the resistor. National Semi used to have a good description of this problem and cure in the LM324 datasheet. Paul {Quote hidden}> I have built a volume control circuit using opamps and a digital pot. > The opamps I am using are NE5532N. The input buffer and output driver amps > are configured as inverting amplifiers. This is so I can add gain later if > need. My problem is that as the circuit reaches unity i.e. > digital pot is at > 0. The input amp has a hi frequency osc. This is not audible nor does it > noticabley affect the audio but should not be there. I have read somewhere > that I should have some capacitance somewhere to stop this but > can not find > the reference :-(( -- http://www.piclist.com hint: To leave the PICList EraseMEpiclist-unsubscribe-requestspam_OUTTakeThisOuTmitvma.mit.edu

Paul Hutchinson wrote: > >In addition to the good suggestions others have made here's something I've >seen often. > >If the load you're driving has too much capacitance you can get oscillation >on the output of most op-amps. With a LM324 type it takes only a few tens of >pF to make the output oscillate. Some op-amps can tolerate a lot more load >capacitance. To decouple the op-amp from the capacitive load simply add a >small (10 to 100 ohm) resistor in series with the output. If the series >resistance causes too much drop in the output you can move the feedback >takeoff from the output pin to the other side of the resistor. National Semi >used to have a good description of this problem and cure in the LM324 >datasheet. > This is quite true. Sometimes you won't even see an oscillation, it just appears as a DC offset from hell (nothing you do will make it go away). I've had to use the resistor several times, with the DC feedback resistor moved to the 'load side' of the series resistor (so that my gain is not affected). Do not move your compensation capacitor though! You're asking for trouble if you do, as you are dropping one of the pole frequencies, reducing phase margin and contributing to the instability. Keep the capacitor connected between the op amps output and the inverting input. Jim Nagy Elm Electronics ICs for Experimenters http://www.elmelectronics.com/ -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Matt wrote... >I have built a volume control circuit using opamps and a digital pot. >The opamps I am using are NE5532N. The input buffer and output driver amps >are configured as inverting amplifiers. This is so I can add gain later if >need. >My problem is that as the circuit reaches unity i.e. digital pot is at >0. The input amp has a hi frequency osc. This is not audible nor does it >noticabley affect the audio but should not be there. You're right; it shouldn't be there. It may not be audible, but it very likely will cause significant distortion since the oscillation will generally vary- sometimes a little, sometimes a lot- with the instantaneous voltage of your audio signal. >I have read somewhere >that I should have some capacitance somewhere to stop this but can not find >the reference :-(( Here are a couple of application notes at the Analog Devices, Inc. website that might help: "An I.C. Amplifier User's Guide to Decoupling, Grounding, and Making Things Go Right for a Change" http://www.analog.com/techsupt/application_notes/AN-202.pdf "Careful Design Tames High Speed Opamps" http://www.analog.com/techsupt/application_notes/AN257.pdf "Analog Signal Handling for High Speed and Accuracy" http://www.analog.com/techsupt/application_notes/AN-342.pdf Based on your description of the problem, my guess (and it's only a guess) is that you might be having problems caused by capacitive loading on the opamp outputs; if that's so, putting a small capacitor between the opamp output and the summing junction of your amplifier (i.e., the opamp's inverting input) won't help matters. For that, you need to isolate the opamp output from the capacitive load with a small series resistance, maybe ten to a hundred ohms. Again, that's a guess; there are lots of things that can make an opamp oscillate. Perhaps the best thing you can do is get yourself a copy of Bob Pease's book, "Troubleshooting Analog Circuits" (US$20.26 at Amazon.com), which is packed with practical information on how to make analog stuff work right. >Also what value input decoupling capacitor do you recomend What's an "input decoupling capacitor"? Do you mean input coupling capacitor? Or power supply DEcoupling capacitor? :) Dave -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics