freestompboxes.org

fakcior wrote:Use of the diodes make different opamps sound more similar, but it doesn't mean they sound the same.

Well, the diodes are compression limiting/clipping or soft clipping to a point, and up to that point they are still retaining much of the input signal that is fed to them from the Opamp. It is not until the peaks square off that Hard clipping begins making different Opamps sound similar. I find the beginning of this threshold to be around the half way spot of the Distortion potentiometer with the Volume potentiometer full on. And with stock cap and resistor values as well as stock 1n270 diodes, this threshold is dictated by the Opamp used.(I demonstrate this below) The point is, the vintage 741CP, UA741CP are different then the LM741CN even though the spec sheets may state they are exactly the same. Yes, other type of single Opamps used within this circuit will sound different, however, the same is true with the different 741's used in the original Script and Box logo'd MXR Distortion+'s. Although you may have to strum something other then an open G chord to actually hear or feel the difference.
(Note: My Favorite's were the Original 74 Script 741CP chip and the LF356)

TL071


LF356


LM741CN - company A


LM741CN - company B


UA741CP


741CP - Original 74 Script chip


~waldo

I may have misread but otherwise,
Could you post the scope settings you're using? Volts/Div for both channels as well as the timebase. (i'll calculate the input freq myself )

And is there a reason for using a rounded triangle as input instead of a sinewave?

Dirk_Hendrik wrote:I may have misread but otherwise,
Could you post the scope settings you're using? Volts/Div for both channels as well as the timebase. (i'll calculate the input freq myself )

And is there a reason for using a rounded triangle as input instead of a sinewave?

rounded triangle? the input signal is a 2.5Khz sine wave at .1 Volts/Div and .1ms timebase. The output for the first opamp only swaps is a 1 Volt/Div and the second opamp swaps with the 1n270 diodes is 50mV Volts/Div.

~waldo

Can you point to the reason why the ouptut waveforms are different for those two op amps, AND, more importantly, is the difference even audible? Can you point to anything in the specs for those op amps that would account for the different clipping behavior?

Do the same test with the scope while keeping the diodes in the circuit. Is the output waveform different for any of the op amps?

waldo041 wrote: rounded triangle? the input signal is a 2.5Khz sine wave at .1 Volts/Div and .1ms timebase. The output for the first opamp only swaps is a 1 Volt/Div and the second opamp swaps with the 1n270 diodes is 50mV Volts/Div.

~waldo

Thanks!
Why did you choose a frequency this high? Considering the guitar use a frequency in the <1kHz range is more realistic.

And btw, your input sine wave is rather distorted

Dirk_Hendrik wrote:I may have misread but otherwise,
Could you post the scope settings you're using? Volts/Div for both channels as well as the timebase. (i'll calculate the input freq myself )

And is there a reason for using a rounded triangle as input instead of a sinewave?

It could be that the input signal is being distorted into a rounded triangle waveform due to the op-amp's slew rate.....

Also, the biasing of the op-amp's inputs might need a bit of tweaking to get it to clip evenly....

Would it be possible to post the schematic you used in the test setup?, I'm interested in breadboarding this so I can check things out......


Note: Apologies for my multiple posts, there doesn't seem to be any edit button for this thread......

Maybe the answer to why different 741 op-amps sound different is probably because they have different slew rates, I seem to remember reading somewhere that the original 741 op-amps weren't particularly stellar with regards to performance, hence the reason why later,more modern op-amps with Fet Inputs were developed.......

DrNomis wrote:Also, the biasing of the op-amp's inputs might need a bit of tweaking to get it to clip evenly....

DrNomis wrote:Maybe the answer to why different 741 op-amps sound different is probably because they have different slew rates, I seem to remember reading somewhere that the original 741 op-amps weren't particularly stellar with regards to performance, hence the reason why later,more modern op-amps with Fet Inputs were developed.......

Excellent points. I wondered about the bias as well. I also missed the 2.5kHz test signal as Dirk pointed out.

I think a comparison of specs could be helpful.

DrNomis wrote: It could be that the input signal is being distorted into a rounded triangle waveform due to the op-amp's slew rate.....

All your other q's are valid. This one appears as odd to me. As far as I interpreted the smaller "sine" wave is the input signal. I do not see how that one could be distorted by the opamps slew rate. Furthermore, if this were the case, when looking at a triangle wave as a series of harmonics (read: Fourier) if would be odd that the opamps poor slew rate, essentially functioning as a filter since the opamp loses "interest" in the signal, would add harmonics to the signal. The reverse would be far more to be expected.

Dirk_Hendrik wrote:

DrNomis wrote: It could be that the input signal is being distorted into a rounded triangle waveform due to the op-amp's slew rate.....

All your other q's are valid. This one appears as odd to me. As far as I interpreted the smaller "sine" wave is the input signal. I do not see how that one could be distorted by the opamps slew rate. Furthermore, if this were the case, when looking at a triangle wave as a series of harmonics (read: Fourier) if would be odd that the opamps poor slew rate, essentially functioning as a filter since the opamp loses "interest" in the signal, would add harmonics to the signal. The reverse would be far more to be expected.

That's a good point, I was looking at the scope screenshots and thinking that the triangle waveform looked alot like what you get with slew rate distortion, it's obvious that something in the circuit is causing the waveform to be shaped like that......


Just had a thought, could loading effects do that?......

Define loading.
resistive, inductive, capacitive, non-linear
If for whatever reason the input of the effect contains a DC component (leaky input cap) that pulls the drive stage of the sine generator off it "could" result in distortion. But all that is really theory. The pics in this thread suggest the sinewave generated is not that great. Most likely it's a filtered triangle with a poorly calibrated filter.

so is the suggestion being made that the slightly distorted sine wave is giving the results shown in the pictures?

Also, oops, I forgot there was a time variable applied and not accounted for and the actual calculated frequency is 1.5khz

~waldo

DrNomis wrote:Also, the biasing of the op-amp's inputs might need a bit of tweaking to get it to clip evenly....

I am attempting to visually show that the two opamps MXR used, the UA741CP/741CP and LM741CN do have differences within this circuit. I am using the same layout for both the script and block logo MXR Distortion+'s, the only changes are a socket(s) for the opamp and diodes. If I were to rebias the opamp to "tweak", I would alter the circuit which would negate the point in showing that the opamps behave differently within this circuit.

~waldo

Not a surprise, here is a 1Khz SINEWAVE

Input = .1V - Volts/div
Output = 50 mV - Volts/div

MXR Distortion+ Circuit, only Opamps swapped in and out.
Volume = Full rotation
Distortion = 45% rotation (+/- 5%)



TL071


LF356


LM741CN-company a


LM741CN-company b


UA741CP


741CP-vintage


~waldo