Hey All,

I've been having fun w/ building distortion circuits for keyboards and guitars. I'm working w/ the LF347 op amp. I've noticed three things:

1) There's still a hint of original signal going through the circuit when the circuit isn't powered. I'm guessing a higher valued resistor on the input would solve this problem. (2.2k, 5k, etc.)

2) There is a distinct hum in the output. I'm thinking of a passive low pass filter for this.

3) I noticed that when turning the volume on the keyboard amp w/ this circuit connected, I noticed there is a sudden jump in volume. And it then slowly progresses as the volume is turned up. This is not the case when the circuit is not connected.

Datasheet on the op amp:

http://amiga.serveftp.net/Datasheets/LF347-QuadOp-amp.pdf

 

Probably unrelated to your observations:

When you build an opamp circuit, unless specifically directed otherwise, you should always put 0.1 uF and 10 uF from each power pin to ground, connecting them as close to the power pins as possible.

Typically, the gain is not varied, to vary the output level. Typically, an amplifier circuit with a fixed gain is used with a resistive divider (potentiometer or stepped attenuator) before the input, possibly with an added unity-gain buffer amp after the divider.

I highly recommend going to http://www.national.com and downloading Application Notes AN-20 and AN-31.

 

1) That input resistor idea won't fix it.
2) Fixing the source of the hum is always better than trying to fix the results.
3) this chip has a failure to operate with its input within 3 or 4 volts of the negative voltage supply.
4) Your "theoretical" drawing is useless for figuring out problems with the real thing. It doesn't show the DC bias point, how the hum is coming in, or how the volume jumping is happening. A good drawing still might not show YOU where these things are happening, but it might show ME. Give us a real drawing and you might get some good advice.

Sorry to be short, but I have a hurry-up to do right now.

ps, you can put the gain control in the feed back loops, but you run into noise problems sometimes.

 

Hello,

The ground connections are floating as they are only connected by the two capacitors C1 and C2.
Connecting a couple of resistors of say 10K accross the C1 and C2 will set the gound to 1/2 the battery voltage.

Bertus

 

Hi, this is my first post on this forum, so here goes:

I have to be a bit negative here (sorry), but there are some problems with your circuit:

1. Your power supply is floating (in a dc sense) relative to ground, which means all manner of silly things might happen (anything but your expected behaviour anyway). If you want to use just one battery, then perhaps the easiest solution would be to operate "above ground". You then need to put dc-blocking capacitors at input and output - not ideal, as you'll get "thumps" at switch-on and off. (If this is a problem, then you'll need a different solution.)

2. The pot R1 shouldn't have one end grounded anyway, as this can result in the op-amp trying to drive into a short-circuit.

3. Relying on an op-amp to overload in a graceful manner isn't the best way of achieving what you desire. Usually, you would use back-to-back diodes across the feedback resistor.

4. Guitars usually need to drive into a high impedance, so you'll need to address this.

My solution for a simple circuit which ought to work reasonably well would be as follows:

Firstly, make use of the non-inverting amplifier configuration (this is the easiest way to get a high input impedance). I strongly suggest you get to know how to adapt this to single supply operation by googling "op-amp single supply". When you're happy, then all you need to do is ensure you put back-to-back diodes across the feedback resistor. And make sure your decoupling/reservoir capacitors are nice and close to your op-amp's power pins.

Sorry to be so negative, but I wish you well with your little project - and if there's anything you don't understand, don't be afraid to ask!

 

I just realised "back-to-back diodes" might be very confusing; what I mean is 2 diodes, each connected across the feedback resistor, with mutually reversed polarity. This will result in fairly "soft" clipping, I would think, in a non-inverting configuration.

 

hi,

it's been quite sometime that i have participated in our forum.

1. is the schem that you have uploaded the real one?

 

hi,

1. is the schem that you have uploaded the real one?

Yes this is the schematic that I'm working w/.

 

I just realised "back-to-back diodes" might be very confusing; what I mean is 2 diodes, each connected across the feedback resistor, with mutually reversed polarity. This will result in fairly "soft" clipping, I would think, in a non-inverting configuration.

Yes, I understand what you are saying about the diodes. I posted this simple schematic because I removed all of the diodes and other items that I added to get down to the crux of the circuit to find that hum I mentioned in my original post.


You had a lot of good things to say, which I will try out later this afternoon.

 

Probably unrelated to your observations:

When you build an opamp circuit, unless specifically directed otherwise, you should always put 0.1 uF and 10 uF from each power pin to ground, connecting them as close to the power pins as possible.

Typically, the gain is not varied, to vary the output level. Typically, an amplifier circuit with a fixed gain is used with a resistive divider (potentiometer or stepped attenuator) before the input, possibly with an added unity-gain buffer amp after the divider.

I highly recommend going to http://www.national.com and downloading Application Notes AN-20 and AN-31.

I liked AN-31. It had a lot of different methods for using op amps that I wasn't familiar w/. A lot of good info in there. I'm going to try what everyone is saying about grounding. I'm thinking the hum is mostly a grounding issue.

 

The hum might be from a cheap wall-wart power supply or it might be mains hum pickup by an unshielded input cable.