NE570 and op-amp Questions

[soggybag]

I built the Tremolo-Matic X from the Stompboxology. It seems to work pretty good. I had the idea to add a few improvements.



1) There is a little noise/distortion when the input is too hot. The noise is either coming from the 570 or the From reading Thomas Henry's book "Making Music with NE570" I have found the following:

The current into the gain cell must be limited to ±140μA maximum on signal peaks. Thus when us-
ing internal resistor R2 only, the voltages on pins 3 or 14 must not exceed ±2.8V.

In the schem above pins 3 and 14 are preceded by a 6K8 reisistor. Note that pins 3 and 14 have an internal 20k resistor built into the 570. This is fed by an op-amp voltage follower U4b and c, and before this the input buffer, U4d. The input buffer has a gain of about 7.

The first question then is how to limit the signal going into the pins 3 and 14 to +-140uA/+-2.8V.

It seems to me that I could lower the value of R14 but this would lower input impedance, which might not be a good idea. Increasing R13 might do it also. But then the question becomes what values to use. So what kind of voltages should I expect from a standard guitar input? With the values shown I figure that Boscorelli was planning for signals up to 3.7v. From what I remember reading a typical electric guitar signal might show an immediate peak in the 5v range.

Maybe R8/15 should be 20K. This plus the internal 20K resistor would put 40K before this input.

2) After looking at this for a little while it seems the inverting input is a little strange. This circuit sounds pretty good, but it looks like the input impedance is a little lower than most designers would like.

It appears this was done because the NE570 has a built in op-amp that is used the output buffer. This op-amp must be used in an inverting configuration. There are two of these op-amps which are accessed through pins 5,6,7 and 10,11,12.

Looking treble path the input is inverting U4d, followed by non-inverting follower U4c and then onto the gain cell of the 570 and to the output buffer op-amp on the 570 which is also inverting. This was done to keep the signal in phase at the output.

After some thought it occurred to me that if the input buffer were non-inverting the follower stage U4c could made inverting the input would stay in phase with the output. This would also allow for a higher input impedance.

My first guess would be to flip the + and - inputs and use two 10K resistors one going before the - input and the other between the - input and the output. Last of all tie the + input to VB.


3) Another thought occurred to me as a possible mod. As it is the circuit basically pans between the bass and treble signals. These are mixed and sent to the output. As it is the bass signal comes from pin 12 of the 570. This is the out put of the gain cell and bypasses the op-amp in this half of the chip. The idea was to use the op-amp at pins 10,11 and 12 to invert the bass signal before mixing it with the treble signal.

The idea then is to mix the out phase treble and bass signals. I'm not sure what this would sound like. But I'll guess that you would get the treble or bass signal at either extreme. While at a 50% mix the out of phase signals would mostly cancel each other out for a more conventional tremolo sound.

The next half of this question is exactly how to wire this up...