Am 12.10.2012 10:15, schrieb Richard Prosser: > OK Alex. > > try > > I've adjusted the gains etc to be close to your needs - some > adjustment may still be required. I set the second input at 15.4V as > this is 100x the 154mV for input1. Thanks Richard. When writing 100x I was talking about the relationship=20 between the signal AFTER the transformer, ie at the Node "Transformer",=20 154mV/6 =3D 25.67 mV. And that is Vpp, the voltage in the simulator is=20 amplitude, so it's 1283mV or roughly 1.2V Amplitude for V3. > Reduce R27 for more input2 gain if > you need it. I changed R27 to 33k and V3 to 1.2 Volts. BTW: I get a flat line at the=20 peak. Does that mean distortion? If so how to increase the gain without=20 hitting the limits? > The output load of 20ohms will be a problem - the opamp probably won't > work too well into that low a load. 6V pk-pk into 20 ohme works out at > about 330mA pk-pk. Basically all my headphones "say" is that 3V is the maximum voltage to=20 be applied. 25 Ohms was a good guess. I just measured the resistance to=20 be 90 Ohms on the left channel and 103 Ohms on the right channel. So as=20 I connect both headphones' inputs, the common resistor should be 48=20 Ohms. Add to that 1 kOhms and you get 65.5mA at 3V. > Also the frequency low end response is limited by the transformer. > You'll need to check the exact parameters of your transformer to get > a good estimate. This ( http://www.mikrocontroller.net/attachment/156735/NFU.pdf ) is all=20 I know about the transformer. > > Re the opamp - apart from the output current, I'm not sure what to > suggest. Add in a proper audio output amp chip perhaps. Or include a > simple 2 or 3 transistor output stage, similar to your earlier input > amplifier. Which circuitry would you add to make the voltage be around 2 - 2.5=20 Volts (resp. 4 - 5 Volts Vpp) without distortion? How do you compensate the offset? > > Hope this assists. Yes it does quite well. Thank you for your time so far. Current schematic: Alex > > Richard > > On 12 October 2012 20:19, Alexander Winnig > wrote: >> Am 12.10.2012 05:34, schrieb Richard Prosser: >>> Ciruit2 should be OK with a few tweaks. >>> >>> But some questions arise also. >>> 1. what is the frequency range of interest? >> 10 Hz to 20 kHz. >>> 2 The desired 3V output - is that pk-pk or rms? >> Vpp =3D 6V. This is the MAX! It shouldn't be reached when the input sign= al >> is at max, it should be around Vpp =3D 4V to 5V then. >>> 3. What are the actual input signal levels? >> Vpp before transformer should be 0.154V or 154mV max. >>> 4. What is the supply? If it is a 9V battery, then your 30ohm biasing >>> resistors are going to drain it fairly quickly. >> Yes I am. >>> Better to use higher >>> value resistors and bypass with 10uF or larger caps. >>> >>> Other Comments: >>> You have the transformer turns ratio set at 1. This isn't in >>> accordance with your 1.6:1 statement above. >> On my transformer there is "1:6, RPrim =3D 4.6 Ohms, RSek =3D 180 Ohms". >> >> Do you mind modifying my circuitry ( >> https://www.circuitlab.com/circuit/ru78q2/zusammenfugen/ ) using your >> ideas and republishing it? Makes this conversation a lot easier. >>> Are you actually going to use a '741? If you are, then you will have >>> limited gain for any reasonable frequency range. And noise may be an >>> issue also. >> I have >> >> LM358 >> LM393 >> 741CN >> LM2901N >> >> at my disposal. Which one would you recommend? Or would you rather >> recommend building an amplifier the way I did in the first schematic ie >> using transistors? >> >> Alex >>> RP >>> >>> >>> >>> On 12 October 2012 16:06, Alexander Winnig >>> wrote: >>>> Both signal sources are being shown at max amplitude. My goal is to mi= x them >>>> so each signal source has the same volume and the final result is not = to >>>> exceed 3 volts. >>>> >>>> Here's the current schematic. >>>> >>>> https://www.circuitlab.com/circuit/x6hjxg/verst2/ >>>> >>>> Feel free to modify and repost the link. >>>> >>>> But I thought that maybe I can save a lot of components and do it like >>>> >>>> https://www.circuitlab.com/circuit/ru78q2/zusammenfugen/ >>>> >>>> Factor between source one and source two is about 100. Thus the resist= or >>>> values R24 and R27 to make them even. But is is not quite satisfactory= .. It >>>> seems that -(V(Transformer)/R27 - V3/R24) * R12 is not -3 Volts as you= 'd >>>> expect from an adding amplifier. Also there's something wrong with the >>>> transformer. It's an 1:6 transformer that should produce about 0.013V = on the >>>> output side. >>>> >>>> I am not sure but I guess it is a better approach to optimize circuit = 2 >>>> rather than the old one. >>>> >>>> Alex >>>> >>>> >>>> Am 12.10.2012 04:05, schrieb Richard Prosser: >>>> >>>> Maybe you meant something more like this? (Minimal Changes and one >>>> input changed to 200Hz to separate the inputs). >>>> >>>> RP >>>> >>>> On 12 October 2012 14:16, Dave Tweed wrote: >>>> >>>> Alexander Winnig wrote: >>>> >>>> I created this https://www.circuitlab.com/circuit/r2kea4/verstarker/ >>>> circuit. You can open it and simulate it in your browser, no additiona= l >>>> software needed btw. >>>> >>>> But simulating creates a weird pattern. Is this really the output or >>>> just some flaw in the simulator? >>>> >>>> What you see is an amplifier on the left side and a signal mixer on th= e >>>> right side. So the result of the first amplifier acts as second source= .. >>>> Measurement is taken at R15. >>>> >>>> What I see is what looks like the ground node of the mixer connected t= o the >>>> output of the amplifier, and the upper input to the mixer connected to= the >>>> amplifier's power supply. Surely this isn't what you intended. >>>> >>>> -- Dave Tweed >>>> -- >>>> http://www.piclist.com PIC/SX FAQ & list archive >>>> View/change your membership options at >>>> http://mailman.mit.edu/mailman/listinfo/piclist >>>> >>>> >>>> >>>> >> -- >> http://www.piclist.com PIC/SX FAQ & list archive >> View/change your membership options at >> http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .