General design theory help - Page 2

been busy with school. but here's a quickie.

Generally, you want to have the input impedance of the non-inverting input and the inverting input to be the same.

Hence, R1 || R2 = R3 || R4.

This applies greatly to BJT input opamps where there is enough base current to cause dc errors.

Quote:

Originally Posted by Avro_Arrow 

To balance + to - input, R1 ll R2 = R3 ll R4

R1 ll R2 = 666.67 ohms

R3 ll R4 = 666.63 ohms

This only applies when there is a DC path through R1 to ground but the cap C1 breaks this path.

Regarding DC coupling the inputs on a BJT input opamp: It is somewhat risky to fall into the trap of thinking that the source represents a return path to ground potential. Although this assumption is usually OK (when the source has DC or transformer coupled coupled outputs) it could land you in a spot of trouble when this assumption fails which will happen with a source with an output cap.

Without the cap the input attenuator (if used) is included in the DC path back to ground so the effective value of R1 varies as you spin the knob which makes life hard again. 

On that note: if you use an input attenuator ~32Kohm is an odd input impedance. Its a little low for even a 5Kohm pot. unless you have a specific reason to want to do it this way R2=90-100K with a 10Kohm pot will keep everyone except people with tube output gear (driving the output directly off the plate) happier. Ideally R1+R2 should be 10* the nominal value of the pot so that it dosnt load the output of the pot, but you can get away with a little less, which buys you some wiggle room in R3&R4 and may be the easiest way to get the impedances to line up.

Maybe not a circuit design question, but more of an application problem. When choosing transformers, is it safe to use a transformer with current rating that's close to the circuit's required current? Or would it be better to use a transformer rated at twice the required current?