Tips for building hi gain amplifiers

[Dajgoro]

Every time I need to build a hi gain amplifier for a mic or something else I always end up having issues with the circuit. Last time I tried I built a dynamic microphone preamp with a balanced input (input transformer). I just stacked two A class amps to get a gain high enough, but then the thing turned out into a oscillator, after a lot of tweaking I managed to reduce the oscillation. The other problem is the power supply. Since nowadays all AC adapters use a switching supply, the noise in unbelievable. The only way I managed to get rid of that noise for the preamp was to use a transformers secondary in series(because I don't have any decent inductors around), and place it as far as possible from the preamp. Another problem is interference, I have to keep the preamp away form any other devices, especially CRT stuff, and I already enclosed the thing in a tin can.

So I would like to hear tips form the experts on how to avoid such issues with better design practices.

[Paul Price]

Do you have a circuit in mind you are working with?

Learning how to make practical working circuits means often to do a little research on the internet to discover how others have mastered their design problems. There is a ton of information and many circuit examples to be found that work well. You need to figure out why these circuits succeed to do their job well..compare your circuit inventions with them.

 Most noise problems can be solved with larger valued electrolytic bypass capacitors and sometimes adding small valued inductors and low value resistors to separate each stage's separate power supply connection point to the main supply voltage.

Add a resistor-electrolytic bypass capacitor between the input stage power connection and the main power supply input point for the output stage will greatly help to prevent output stages from feeding back to the input and creating oscillation.

Even electronic design experts spend a good deal of design time and have to try many approaches to circuitry to prevent their oscillators from just amplifying and their amplifiers from just oscillating.

[tszaboo]

If you have bandwidth, offset and other problems in high gain amplifiers, you should use more stages for the amplification. There is usually no other way, as you cannot just increase the GBW of an amplifier. Oscillation can be caused if the Phase shift of an amp is close to 180 degrees, which means we need more speed for it.

For the power supply: never trust an AC adapter. Use linear regulator, switching regulator, anything after that, but dont trust that. The cable will drop votlage, pick up noise, different voltages will come out of it, it has too much problems, to ignore it. Use something which is characterized correctly like a western (or Japanese) power supply IC and support circuits.

[Dajgoro]

I did actually add a 7805 to the power supply, but that didn't help much. The resistor/capacitor array helps a bit, but that ain't enough. A good inductor solves a good deal of the interference, but the problem is that the interference gets transmitted to the input by the proximity of the two devices. I do have the input and output set far away, but that still isn't enough. I don't use decent pcb, because I lack the tools to make one, so I just use point to point soldering on perfboards.
When it comes to using existing schematics, I didn't have much luck with those, especially when it included opamps, also when using opamps like the LM358 I get too much noise.
Another thing, I got the TL072, because the datasheet states "High Input Impedance . . . JFET Input Stage", but when I built my preamp, it didn't work at all, when I connected the voltmeter to the input pins I was measuring like 7V, and when I power the opamp with all of the other pins disconnected, and when I measure the voltage on the input pins I still do get a random voltage? What is going on here?

[Paul Price]

Draw your circuit schematics, just freehand or use a schematic capture program and scan it into your PC and post it here..will make any explanation much clearer, since it is too difficult to try to guess exactly what you are saying just from a description.

Working with electronics involves some luck, but so much more it involves careful planing and evaluation and thought.

[Dajgoro]

I've tried lots of various schematic, what I am looking here are tips, and things to watch out for that might cause problems.

[KJDS]

You need to include the power rails and ground as real elements in your design. If there is a path from the output stage to the input stage via power and ground lines then that's an additional source of feedback. Add in some decoupling capacitors that aren't perfect, and aren't perfectly grounded and they will have resonances that provide phase shifts that can easily cause oscillations, usually at a frequency far higher than the intended operating frequency.

[GK]

Use gain blocks that have good power supply rejection. Discrete, ac-coupled common-emitter gain stages are a potential recipe for disaster when cascaded in duplication.

[ampdoctor]

As others have said, without seeing the actual design I can't make any comments other than very general suggestions. And when you say high gain designs I'm going to assume you're talking about something along the lines of 65-70 dB.  A few things I might want to add to what's already been mentioned are first and foremost pay close attention to grounding and power supply decoupling, and I mean pay REALLY close attention! Even when you're sure it's perfect you still may have issues and require some changes. Also, good shielding is a must, both within the circuit itself and from the enclosure to the outside world.  Another one is that while using opamps is hardly a deal breaker, you might be better served by going with a discrete bjt design with transformer balanced inputs. I suspect, and I could be wrong, that these topologies dominate high end pro audio not because they have some affinity for old technology, but rather the components will be inherently lower noise, and when laying out the board you can optimize component placement to minimize crosstalk, parasitic capacitances, etc.

[azi]

Maybe I misunderstood, but it looks like your basic design is just wrong. I suggest you go to the THAT Corp. website (http://www.that.com) and look at their design notes and datasheets for their mic preamp products (1510 and 1512). You can learn a lot about how to design a good mic preamp there.

Check with the other manufacturers as well. Research mic preamp chips at Mouser, Digi-Key or whatever, then find the manufacturers' sites and look for app notes.