Rat Impedance/Chip Question

[Premium.Dirtbag]

Evening All,

First post so I'd like to just note that I'm brand new to this with little to no previous knowledge on what I'm about to ask. Everything is self learned over the last few weeks if its not evident enough so thanks ahead for your patience. Super excited to keep learning more.

I originally decided to pass the quaran-time with two BYOC kits which I did fast. Ended up gleaning enough to complete a Rat and Distortion + breadboard this weekend entirely from the Electrosmash breakdowns. Now that I'm done, I'm curious to dig into the info they present to better understand what I did and can do. As such, I kinda went down an IC rabbit hole today and am trying to basically understand the numbers I'm seeing versus the sounds I'm hearing.

I ran a 5534, a TL071, and a 741 through the same build. I recorded them into Ableton proper to find that the 5534 had more top end and a higher frequency mid push that I liked more. It also cleaned up the nicest and was the brightest when rolled all the way back. The 741 was my least favorite. It sounded good full out but was missing a little bit that I feel would go along way in a live context. The clean up sucked, it stayed dirty and was muffled as evidenced by the lack of top end when fully dimed compared to the 5534. The TL071 was a few dBs louder and had something the 741 didn't but it still wasn't close to the 5534.

I say all of that to ask why? I understand the slew rates of the TL071 and 5534 are comprable with the 741 being much lower. Besides the slew rate, looking at whats going into the circuit up to that point, the only real changes I notice are the input impedances. If I understand impedance correctly (I've ran the numbers based on the Electrosmash equations) the T071 and 741, being closer to 499k Zin/400k Zin respectively should retain a higher top end than the 84k Zin of the 5534. The voltage loss on what I'm crunching between the first two and the 5534 is around -0.6 more on the subsequent math. Shouldn't that mean a lower resonant peak on the 5534?

Honestly math was never my strong suit and am suspect of my understanding thus far so I'm sure someone smarter can help, really appreciate it.

Also, I'm also kinda planning to box it up eventually and since I have the output coming right off the 100k pot, I was curious if it's advisable to buffer that down or if 100k to 10k is really negligible once you've chained it up in a board?

[antonis]

Hi and Welcome..
(still waiting for Aron's "Welcome button" implementation..)

Can't follow your maths for input impedance but 84k for a 100k 5534 bias resistor seems to be fair..
(100k value is also fair for bipolar op-amp bias..)

But, for input HPF cut-off frequency, 100k bias resistor value is what counts, so -3db cut-off point is raised at 72Hz (from 7.2Hz..)
Of course, by raising input coupling cap value 10 times (220nF from 22nF) you come back to "original" HPF corner frequency..

[rankot]

You may try to use 741 with an input buffer which will add more input impedance.

[Premium.Dirtbag]

Hi and Welcome..
(still waiting for Aron's "Welcome button" implementation..)

Can't follow your maths for input impedance but 84k for a 100k 5534 bias resistor seems to be fair..
(100k value is also fair for bipolar op-amp bias..)

But, for input HPF cut-off frequency, 100k bias resistor value is what counts, so -3db cut-off point is raised at 72Hz (from 7.2Hz..)
Of course, by raising input coupling cap value 10 times (220nF from 22nF) you come back to "original" HPF corner frequency..

Apologies if the coffee isn't quite working its magic yet but let me see if I follow. The Zin of the chip may not have as much importance as I figured originally? The biasing resistor/input cap will do more for the high frequencies in relation to the 5534 than any of the math I was obsessing over? I'm also eyeing R5/C6 to help tighten up some flub.

I guess where I got hung up was the fact the Zin and Zout of the circuit are so low/high respectively at 84k/100k. Everywhere I read says try to aim for 1M in, 10k out (within reason of course). Am I worrying about nothing...if it sounds good it is good etc or would I benefit from buffers?

You may try to use 741 with an input buffer which will add more input impedance.

In theory, it would for any of the chips right? At this point, I'm working under the assumption that the 741 being dark and muddy compared to the 5534 has something do with the 0.5v Slew Rate of the 741 being way under the 5534's (or does the comp cap fix this).

That said, the Zin of the 741 is already almost 5x that of the 5534 so putting an input buffer on it for more highs begs the same question haha. I'm guess I'm just way curious to know why the 5534 sounds "sweeter" and cleans up better with way less gain/more highs when it only has an 84k Zin when the 741 has 400k.

(I think I'm showing my naivete but appreciate it guys haha)

[antonis]

The Zin of the chip may not have as much importance as I figured originally? The biasing resistor/input cap will do more for the high frequencies in relation to the 5534 than any of the math I was obsessing over?

Quite right..

Op-amps input impedance is strongly dominated by bias circuit (non-inverting configuration) or even more by feedback loop gain resistor (inverting configuration)..
Bipolar inputs op-amps exhibit significally lower input impedance than Fet ones but, in any case, bias network causes the "issue"..
(lower input impedance needs lower bias resistor(s) due to higher input bias current - you can easily undestand it by the ratio of input/bias values..)
So, in any case, just ignore IC's input impedance itself..

Am I worrying about nothing...if it sounds good it is good etc or would I benefit from buffers?

For non-inverting configuration (even for BJT op-amps) I shouldn't bother at all..

[Premium.Dirtbag]

Does that mean inverting circuits should have more attention paid to that; and why?

Cheers.

The Zin of the chip may not have as much importance as I figured originally? The biasing resistor/input cap will do more for the high frequencies in relation to the 5534 than any of the math I was obsessing over?

Quite right..

Op-amps input impedance is strongly dominated by bias circuit (non-inverting configuration) or even more by feedback loop gain resistor (inverting configuration)..
Bipolar inputs op-amps exhibit significally lower input impedance than Fet ones but, in any case, bias network causes the "issue"..
(lower input impedance needs lower bias resistor(s) due to higher input bias current - you can easily undestand it by the ratio of input/bias values..)
So, in any case, just ignore IC's input impedance itself..

Am I worrying about nothing...if it sounds good it is good etc or would I benefit from buffers?

For non-inverting configuration (even for BJT op-amps) I shouldn't bother at all..

[iainpunk]

hey, welcome to the forum,

this reminds me of the Rat test i did years ago, i concluded that the 741 was really bright and the 071 sounded less "punchy" but in the years after that, i found out that adequate circuit design trumps "mojo parts" like special chips (like the coveted LM308 or the bumble bee capacitors in old guitars) if you want more brightness, take out low end, if you want it to sound "bigger" scoop some mids, but these bits of knowledge, you'll learn along the path of pedal building, they come with experience

cheers,
Iain

[ElectricDruid]

Welcome!

Does that mean inverting circuits should have more attention paid to that; and why?

Yes, you need to pay more attention when you're designing an inverting op-amp stage because the input resistor goes to a virtual ground node, so the input impedance is basically determined by the value of that resistor. If you make an inverting buffer with two 10K resistors (one input resistor, one feedback resistor) then you've only got 10K input impedance. This means you often finish up needing much larger caps to avoid losing bass when you work with inverting stages.

Inverting stages do have advantages of their own though, like the ability to be used as mixers, and the possibility of "gains" below one (e.g. attenuation). You can't do that with a non-inverting op-amp.

It's not super-difficult or anything, but instead just what you said: one has to pay a little more attention.

HTH,
Tom