Help me understand impedance!

[syndromet]

I recently built a sort of a preamp to use between my instruments and microphones and  my computer when recording. It's a simple LT071 running on a +\- 18V supply. I have 10 Uf capacitors both at input and output.

The problem is that when I run my guitar through it (Les Paul with two humbuckers) it sounds ultra-muddy, even at the neck pickup. I guess this has to do with the impedance a the input isn't right? What do I have to do to get the impedance so that my guitar will sound like it should? Is there anyway I can have an impedance that matches both the guitar and the microphone, as it is connected to the same preamp (although through different jacks)?

I appreciate any answer I get. Thanks a lot!

[pjwhite]

Let's see a schematic of your circuit.  That should make it easy to answer your question.

[petemoore]

  10x rule: You want the impedance of the next thing [active stage, often connected by cables] to be larger than the output impedance driving it.
  As I understand it that's a pretty good rule, otherwise the output>input match could run out of uumph. How much power does it take to drive the input, does that load the output down ?
  One way to bump impedance up to a point where it doesn't matter is add a buffer to increase current drive. That might change the voice/sound, by not having say the HF's rolloff as when the signal had a load on it, may or may not be what it ''should be''..that depends on...
  If you have a FF [low input impedance] try a commercial pedal [with the momentary type switching = buffers], note how the clearer highs make sharper fuzz.

[syndromet]

Well, it's really just as simple as mentioned above; a 10 uf on the input, a 10 uf at the output and a pot between the two "gain-pins". No resistors or anything. I remember reading somewhere that adding resistors at the input increases the impedance. Is that true? I'll try it with a bypassed, buffered pedal in front to see if that helps. Thanks for the info, Pete! From what I understand, it is possible to use both the microphone and guitars with the same preamp, as long as the input impedance is higher than the output impedance of the guitar or the microphone?

[valdiorn]

dude, you really don't understand the concept of impedance, do you? 

A huge capacitor like that doesn't act like a very low impedance, so low it just looks like a straight wire connector to the signal. Now,where is that signal going AFTER it has gone through the cap?? A schematic really would help, mostly because there is no such thing as "gain pins" on a TL071 and I really need to know just what you are talking about in order to help you

If you are absolutely unable to post a schematic then at least answer this question: Is the input signal (after going through the capacitor, I presume) going to the plus or minus input on the Op-amp ?? (inverting or non-inverting inputs), and between which pins are you *actually* connecting that potentiometer?

[slideman82]

Why don't you try a 47n cap at the input?

[sean k]

I built a mixer years ago which I wanted to use with guitars, microphones and line level inputs and the simple answer is you can't build a single stage that likes all of them. The guitar and line levels share a 6mm input but it's switched between a voltage amp stage with a high input impedance for the guitar and bypassing that first stage for the line level into what would be termed a buffer.

So impedance, with the caveat that I'm still trying to understand it and am after meaning of the mysterious "j", is how an AC voltage and current come together and are affected by the big three of resistance, capacitance and inductance.

A guitar string waves about in a magnets "magnetic" field and the changing patterns that result in these field changes induce a current in the coils of wire wound around the magnet. Okay, these aren't the proper words to use but it illustrates the mechanical process by which the voltage and current are created. thinner strings over weaker magnets and thinner windings are at a high impedance. Very small voltages and very weak current are created. Using bass strings over stronger magnets, because the string has more mass and is more resitant to magnetic pull, and more windings creates stronger voltages and more current so the impedance is lower, less resistant which is why impedance is measured as a resistance.

Then with a microphone, dynamic, which is mechanical again but the windings are moved, by air pressure, around a magnet the action of creating volage and current in the windings is more "dynamic" than the guilt by association of strings moving in a magnetic field as the guitar pickup and so the voltage may be a little higher but the current is way stronger and so the impedance is low.

It's all quite pyhsical really. Bass takes more energy to create and as a result is at a lower impedance while treble takes less energy to create and decays more easily. combine this with how the energy is transformed into electrical energy and you start to get the idea about what impedance is. High jangly guitar and bassy vocals into a microphone are going to be a total impedance mismatch but bass guitar and high jangly vocals into a dynamic mic may very well match up quite nicely in voltage and current but how those frequencies respond to an input circuit will be somewhat different simply because frequencies act in different ways even when all else is equal.

Even higher in impedance are condensor or capacitor mics which have almost zero current are at best very small voltages and so need amps right on the diaphrapm and DC voltages to support them. Even higher in impedance are ribbon mics which have no DC voltage support an rely on voltage being created in a coil through an extremely light weight aluminium ribbon.

I know I've gone on a bit but can you see the pyhsicality behind the idea of impedance?

acrobats and springy takeoffs require high walls to look their best... the tricks of treble. 

[petemoore]

  Look at the data sheet for Imp.
  Notice certain types of say Jfet Vs. Bipolar can have different input and output impedances, depends on what schematic they fit in...
  There are various ways to increase/decrease impedances, even with a given component.

[Vitrolin]

if your pre-amp works with your microfone, then i (in your place) would go and buy a DI-Box or Direct Box, it can also be made, its basically a transfomer, that maches impedance of a magnetic pick-up, somewhere around 50KHz or so, to that of a mic input, 100-200hz (DI-boxes are used in studios to record bass-guitars).
or better yet connect your guitar to an amp and mic it.

The problem is that when I run my guitar through it (Les Paul with two humbuckers) it sounds ultra-muddy, even at the neck pickup.

by the way the neck pick-up is more bassy than the bridge pick up.

Edit: heres a schematic from jensen:
http://www.jensen-transformers.com/as/as066.pdf

[R.G.]

I have 10 Uf capacitors both at input and output.

The problem is that when I run my guitar through it (Les Paul with two humbuckers) it sounds ultra-muddy, even at the neck pickup. I guess this has to do with the impedance a the input isn't right?

First, as noted before, simply putting a big capacitor in series does not lower the impedance of whatever is already there.  Second your guess is not necessarily correct. Treble loss may have to do with loading, but that's not the only thing that can do it. The circuit itself may have a really bad treble response, or may have a defect that makes it have a bad treble response.

[syndromet]

Thank you so much for all the replays! I'm finally getting a hold on what impedance is, and how it affects the sound.
I had a few minutes to open the preamp, and draw a quick schematic in ms paint. Here it is:


What can I do with this circuit to make it sound good both with guitars and a microphone?

[valdiorn]

yeah that's not gonna work, mate....
Think of the capacitors as a direct wire for the signal (you want caps that are large enough that they allow ALL the signal to pass through it, or else you will get a bass cut, so from the signals point of view, the cap is just like a wire)

What happens when you set the potentiometer all the way to the left?? The op-amp tries to equalize the voltages between output and inverting input, but it will never succeed because you don't have any resistance at the input. Remember that the gain for an inverting op-amp is R_feedback / R_input, but since the input resistor is ZERO (since you don't have one ), you get infinite gain, so that opamp is going clip like a crazy person!

Now, what happens when you set the potentiometer all the way to the right side?? You get zero resistance between the output and input! You now have a direct path between input and output and the op-amp isn't doing squat...

I'm not going to go into detail about op-amps but read this: http://www.allaboutcircuits.com/vol_3/chpt_8/index.html
It's obvious you need a litle more schooling on the subject, so read up before you continue, speanding time understanding and studying a problem is easier, more productive, and much less tedious than doing trial-and-error or hack fixes you don't fully understand, it's also the only way you'll ever be able to replicate your desired results.

[tempus]

Try this:



This sets the input impedance at about 100K which should be OK for your guitar. The gain is roughly the maximum value of the pot (also called the feedback resistor) divided by the input impedance: 500/100 = 5. So you have a maximum gain of 5 which should be just about right for your soundcard input. You can experiment with different values of R_in (the input resistor). The higher you go, the less loading you will have, but the gain will be lower because of the R_f / R_in gain equation. What is the value of the pot you're using now?

Feel free to keep asking questions.

[R.G.]

Thank you so much for all the replays! I'm finally getting a hold on what impedance is, and how it affects the sound.
I had a few minutes to open the preamp, and draw a quick schematic in ms paint. Here it is:

What can I do with this circuit to make it sound good both with guitars and a microphone?

Put a resistor in series with that input cap, and tie the end of the cap that's now being used for input to ground.

Now add a new input capacitor going to the (+) input, and a resistor from the +input to the middle of your power supply voltage.

Now that I see your schematic, your question was not about impedance at all - it was about opamp circuits and feedback.

The opamp circuit you're trying to use is an inverting input opamp. Feedback on an opamp forces the inverting (-) input to be at a very low, almost ground voltage and very low impedance. So your circuit is heavily loading your guitar input through the input capacitor.  What I told you to do was to change the circuit to a non-inverting opamp design. The non-inverting circuit has a very high input impedance, about equal to that single biasing resistor I told you to connect. Now it will not load the guitar if your power supplies are right and the bias resistor goes to a voltage source in the middle of your power supplies.

[CynicalMan]

Try this:



This sets the input impedance at about 100K which should be OK for your guitar. The gain is roughly the maximum value of the pot (also called the feedback resistor) divided by the input impedance: 500/100 = 5. So you have a maximum gain of 5 which should be just about right for your soundcard input. You can experiment with different values of R_in (the input resistor). The higher you go, the less loading you will have, but the gain will be lower because of the R_f / R_in gain equation. What is the value of the pot you're using now?

Feel free to keep asking questions.

100K input impedance is too low for a guitar. It will reduce high frequencies by quite a bit. A useful rule of thumb is to never design clean circuits like this with less than 1M input impedance.

[tempus]

100K input impedance is too low for a guitar

Maybe. I suppose it depends on the pickup, but I don't think it'd be much higher than 20K if that high. I picked 100K as a starting point because, although 1M is probably a better choice, I think the OP will have a hard time finding a 5M volume pot.

Not to mention the thermal noise associated with those high resistances...

[R.G.]

... which is why I suggested changing to non-inverting. That lets you use lower impedances on the inverting input and side step the thermal noise issues except for the ++ side bias resistor.