Trying to figure out input impedance, and how it affects coupling cap value choice

[mzy12]

Hello!

I've been trying on and off over the past year or so to really try and nail this concept down and it has kind of driven me insane

I'm looking for reliable information regarding how you should calculate the input impedance of a guitar pedal circuit and more specifically, whether or not a pull down resistor effects input impedance. I'm also looking for how you should change the input coupling cap's value to accommodate for any change in input impedance in order to get the desired high pass filtering effect for a given circuit.

I've seen numerous resources contradict each other in relation to the pulldown resistor affecting input impedance. For example AMZ over at http://www.muzique.com/news/pulldown-resistor-vs-input-impedance/ says that it does, kind of. If you look at aion fx's documentation for their klon clone, they say that their added pulldown resistor is 2M and R2 (their voltage divider for the opamp) is inceased to 2M in order to maintain the input impedance. PedalPCB's kliche also adds a pulldown resistor, this time at 1M and the R2 value remains unchanged from the original at 1M.

I have a couple of questions about this. One is, is the input impedance of the PedalPCB's klon clone half that of aion's? Surely aion or PedalPCB would have to change the value of C1, the first coupling cap, in order to maintain the same frequency response? Or am I misunderstanding the effect input impedance has on the highpass at the input stage?

Thanks in advance for any response and I hope I posted this thread under the right section in the forums.

 

[damianvila]

See if this helps you in any way: https://www.soundonsound.com/techniques/understanding-impedance

 

[mzy12]

See if this helps you in any way: https://www.soundonsound.com/techniques/understanding-impedance

Checked it out and while it is helpful for understanding impedance when it relates to interconnecting pieces of equipment in a variety of professional and non professional applications, it unfortunately it doesn't contain the information I'm looking for. I think the best way to describe what it is I'm looking for is that I want an overview of the micro level of circuits, like how individual resistors, transistors, op amps and capacitors affect the input impedance and frequency characteristics, where as that sound on sound article is only concerned with the macro level of putting pieces of gear together where the impedance is already known.

Thanks anyways!

 

[finebyfine]

I've used this method to measure input impedance : https://learnabout-electronics.org/ac_theory/impedance73.php, albeit only in LTSpice, but my measurements agreed with the book I was testing circuits from.

Actually calculating it from scratch is far beyond my depth.

 

[BuddytheReow]

Here's what Dr. Bones suggests. With many people here, impedance is a bit over my head, though I do understand the concept of it.

What's all this Impedance Stuff About?

I'll keep the math to a bare minimum, but you can't describe impedance without a little math. When I was in college, the professor in an introductory course on circuit theory shared this anecdote: "As part of their oral exam, PhD electrical engineering students were asked to define the term...

forum.pedalpcb.com

 

[jwin615]

This is why most pedals strive for low output/high input impedance.
1M vs 2Mohm doesn't matter and won't filter anything(in the guitar/audio)in that case.

In a not that correct way of saying it.
If I'm throwing cuos of water at you, and you have buckets to catch it, we're good.
Ohms law. Voltage isn't changing(kinda) from one pedal to the next. It's just the relationship of I(current) and R(resistance).
The only thing left to consider is attenuation and noise but gain structure on a pedal board is, well, kinda dumb. We turn up and down and clip all the time.
The biggest thing, imo, to worry about is the first impedance match, and your cables.
You shouldn't have any real frequency roll offs in a rig unless it's designed into it OR the capacitance of you cable intereacting with the input impedance of another device.

Now, vintage pedals(and pickups) are a little different. The old "fuzz first" has some truth to it, in certain circuits.
But mostly, well, 1M and 2M input shouldn't make a damn but of difference because the incoming signal should be way below that.

Maybe @Amplified Parts could chime in on the vintage fuzz impedance conundrum?

 

[andy-h-h]

Hey there.

I'm also lazy and use LTspice to check impedance. In pretty much every case that I can remember, adding a pulldown resistor to a circuit that doesn't have one reduces the input impedance. It's adding parallel resistance, so it has to come down. The question is, does it actually matter? In most cases no, it doesn't matter that much IF the input impedance is already quite high, or at least high enough to not matter.

Does it affect the frequency response - again, mostly not much at all.

Example: Boss input buffers are around 470k. Ibanez is not that different for the TS. They do the job, and are half the impedance of the somewhat arbitrary figure of 1M that people aim for.

I mostly don't use pull-down resistors, as I earth the circuit input on the stomp switch in bypass, and that seems to have taken care of my needs (at least so far)

 

[mzy12]

I've gotten around to figuring out how to do some spice stuff as some of ye have suggested and yeah, it would appear that the pulldown resistor does indeed have an effect on input impedance. It acts in parallel to the resistor across the 4.5v supply on many circuits. Stupid question - does it matter if you increase the resistance to 2M2 there? I imagine it doesn't, as (I think) it's just a current limiting resistor for the bias of a bipolar opamp.

 

[Amplified Parts]

Maybe @Amplified Parts could chime in on the vintage fuzz impedance conundrum?

I'm repeating some of what has already been stated here, but to be able to connect them in whatever order you'd like, pedals should ideally be designed with high input impedance, and modern ones pretty much always are. Having a low input impedance loads the previous stage more. Low output impedance stages can drive a low impedance load better than high output impedance stages.

High output impedance into low input impedance is when the problem will be the largest. With a low input impedance, a low output impedance source would be able to provide a large amount of current to the following stage. However, with a high output impedance, the amount of current that the output can provide is severely limited, and much of the voltage will drop across the output impedance.

In layman's terms, it causes unnecessary signal loss. Electric guitars themselves have an output impedance that is quite high due to the resistance of the potentiometers and the impedance of the pickups themselves. It's often the first stage following your guitar which is most susceptible to these issues. And on the other hand, many vintage fuzz pedals (like an Arbiter Fuzz Face) have input impedances which are fairly low. This is what we normally try to avoid putting after a high output impedance stage, and yet it's typically recommended that you put something like a Fuzz Face directly after your (high output impedance) guitar rather than after some other effect with lower output impedance.

The reason for this is that the Fuzz Face was almost certainly designed specifically to go between your guitar and your amp. There were very few effects available at the time so big pedalboards with buffered effects weren't around. It expects its fairly low input impedance to be driven by a high impedance guitar output. Doing so causes the output impedance of the previous stage to limit the amount of current that will flow into the base of Q1, which limits the collector current, which lowers the voltage gain, which raises the clipping threshold.

You can drive a Fuzz Face and similar circuits with a low impedance source, but it was most likely not originally designed for that and it will sound different (to a lot of people, worse). There are various ways of approximating the guitar output for when a FF is placed after something like a buffer. Jack Orman's pickup sim is a fairly common method people use. I believe the EQD Eruptor uses something similar:

Guitar Pickups Simulation

Schematics & circuits for guitar effects with vacuum tubes and solid state electronics. FREE downloads of schematics such as TS-9, Fuzz Face and other classic rock devices.

www.muzique.com

You lose the ability to use your guitar's volume knob as an effective fuzz control when doing this, which a lot of people like doing with FFs in particular.

 

[psb962]

....

You can drive a Fuzz Face and similar circuits with a low impedance source, but it was most likely not originally designed for that and it will sound different (to a lot of people, worse).

Am I correct in thinking this would also apply to the Rangemaster circuit? I'm finding it hard to use one of these on my pedalboard as it doesn't sound right when sitting after the tuner (buffered) but when used standalone it's noisy.

 

[Amplified Parts]

Yes it's definitely true of the Rangemaster as well.

Basically, anything with a common emitter stage (input to base, output from collector) at the input will have a fairly low input impedance if it does not have an emitter resistor (e.g. Fuzz Face, Tonebender Mk2, Fuzzrite, etc.) or if it has an emitter resistor which is bypassed for AC by a capacitor in parallel (e.g. Rangemaster, probably others I can't think of right now). If the common emitter amp has an emitter resistor which is not bypassed (e.g. LPB-1), the input impedance will be higher. Larger emitter resistors result in a higher input impedance (when not bypassed). The input impedance of the transistor itself will be approximately hfe*Re, but of course it's in parallel with the biasing resistors and anything else on the input.

How are you powering your Rangemaster? It can inherently be a little noisy and it can depend on the transistor used. But my first troubleshooting step is to try battery or at least a very well-regulated supply to rule out a noisy PSU as the culprit.

 

[MBFX]

If the common emitter amp has an emitter resistor which is not bypassed (e.g. LPB-1), the input impedance will be higher. Larger emitter resistors result in a higher input impedance (when not bypassed). The input impedance of the transistor itself will be approximately hfe*Re, but of course it's in parallel with the biasing resistors and anything else on the input.

For the LPB-1 schematic I am using, the emitter resistor is 390R to ground. Can I simply raise this value to increase input impedance? I suspect this will affect the circuit's performance (amount of gain available?) but I don't know enough yet to figure out how.

 

[Feral Feline]

For the LPB-1 schematic I am using, the emitter resistor is 390R to ground. Can I simply raise this value to increase input impedance? I suspect this will affect the circuit's performance (amount of gain available?) but I don't know enough yet to figure out how.

Raising the value from 390R will lower the gain.

Or, if you mistakenly put in a 390k instead of 390Ω, there will be no gain at all because the transistor won't switch on at all — don't ask me how I know this, nor should you ask about when copying schematics by hand and failing to make sure you don't put a "k" where an "r" should go...

All the circuits are a web — you touch one part and BAM! The giant spider will catch your signal and eat all your tone.

 

[psb962]

Yes it's definitely true of the Rangemaster as well.

...

How are you powering your Rangemaster? It can inherently be a little noisy and it can depend on the transistor used. But my first troubleshooting step is to try battery or at least a very well-regulated supply to rule out a noisy PSU as the culprit.

It's the AionFX kit so it has a charge pump as part of a well filtered power supply, including a pair of 100n caps to kill HF noise. There is nothing wrong with the kit, it's my home network - I feed my router output into the mains then out again to save using spotty wifi or a long network cable run. That system creates 45Hz clicking / digital noise and was driving me nuts until I found it. But that's the only pedal I have that picks it up.

 

[Amplified Parts]

Yes, when it's not bypassed, increasing the emitter resistor with no other changes will increase the input impedance, amp linearity, and bias stabilization for varying transistor specs, but it will decrease voltage gain. It will also raise the collector voltage, so it's probably wise to also tweak the base voltage divider to re-bias if you do that.

With an emitter resistor, the voltage gain is approximately -Rc/Re (negative because it's an inverting amp), so a stock LPB-1 has a gain of ~26 with 10KΩ and 390Ω for Rc and Re. You can compensate the gain decrease when you increase Re by using ~Re*26 for the collector resistor, but that also increases the output impedance and the effects of thermal noise (which is probably not a major concern here but worth noting). With increased Rc and Re, less current needs to flow through the collector to drop the same voltage. So it may need a re-bias, but the increased Re will help achieve less collector current, so it also may be totally fine without re-biasing.