input buffers.....what are they for?

[plexi12000]

hey...i was checking out some pedal schematics, etc.  and this one has an 'input buffer'.   

can someone explain to a big dummy what that does? -lol   thank you

[hymenoptera]

A buffer separates two blocks of circuitry so they have as little impact on one another as possible. This means a buffer typically has high input impedance and low output impedance. Buffers can include transistors, opamps, and transformers, and maybe some others I'm forgetting.

Sometimes you don't want the next part of a circuit to have an adverse or unpredictable effect on the previous stage. This is why you might want an input buffer.

Sometimes it can be desirable (in a euphonic way) to not have a buffer, for instance, guitars typically have relatively high output impedance, a result of the thousands of windings that make up the pickup. Some classic pedals rely on that fact to generate the tone they were known for, and need to be the first pedal after the guitar in order to take advantage of the guitar's high output impedance.

Things to study up on; impedance, voltage dividers, reactance (and the related resistance, capacitance, and inductance), and filters. These are all related to impedance, and will help you get a better grasp on buffers and why they're important.

[plexi12000]

hey thanks buddy-- i really appreciate the comeback.  yes...i do have a lot to study! no doubt there.  it's hard when you just go in blind and no "guidance". lol

This forrum is very advanced but I can pick up tidbits here and there, so that helps.

If you dont mind me asking, today i was looking at the Cornish G2 ODrive.  that also has an input buffer.  There was also a schemeatic w/o the buffer.

my main guitar is a les paul.  7.5k neck p'up and 8.5k bridge.  If i built a G2....should i include the buffer?   thank you

[midwayfair]

A schematic of the pedal you're asking about would help.

Bipolar junction transistors tend to have low input impedance. BJTs also lose input impedance as the gain goes up if they're being used as common emitter amplifiers.

Low input impedance will load pickups, and if a buffer (and I think it's important to mention here that ANY active circuitry is a buffer) is placed between a pedal with low input impedance and the guitar will remove the pickup loading. Because loading decreases gain and deemphasizes treble frequencies sooner than low frequencies, pickup loading can make the guitar sound darker. SOMETIMES THIS IS A GOOD THING. With distortion, filtering high frequencies can make the distortion sound better. SOMETIMES this is a bad thing and can make a fuzz sound harsh.

A G2 is a Big Muff with an input buffer and HUGE feedback treble cuts. The original big muff does load the guitar, but it also cuts a lot of treble in each stage. The G2 is just more of the same. And the BMP tone control was nerfed to be only a treble cut. The addition of a buffer probably doesn't result in a harsh sound here because the pedal was designed to sound like someone was playing under water. (Sorry, my bias is showing ...) But the G2 was also designed to use Cornish's buffer. It's possible it will sound worse without it. Throw it on the breadboard and use your ears.

In this case, one of the other concerns about a buffer, interaction with the guitar knob, doesn't need to be dealt with, because a Big Muff really doesn't react to the guitar pickups anyway. This does appear to be a concern with certain op amp circuits. A Tube Screamer works differently when the input buffer is removed. Worse, in my opinion.

By the way, the numbers you posted for your pickups is just the resistance across the coils. It doesn't necessarily tell you much of anything at all. It's better to use your ears in the interim while you're studying up on hymenoptera's list.

[hymenoptera]

plexi, I don't know if you like to read, everyone learns different, I'm a "reader", but I cannot recommend enough a book by Horowitz and Hill, The Art of Electronics. Everyone doing pedals should have this book if you're interested at all in how these circuits (or anything electrical) actually work. You can find it cheap used on Amazon, ebay, or whatever, or you can have your bookstore order it. It's a reference that you will keep coming back to over and over again as you delve into each new field of study, or just to brush up or clarify something you've forgotten. Seriously, check it out! You're already a step ahead by being curious about how things actually work.

Lately I've been reading the chapter on digital because it's something I never really understood. If you'd have told me a few months ago that I'd have CMOS and 74-series TTL chips on my breadboard experimenting with logic today, I'd have laughed, because I could never wrap my head around that digital stuff, but with the help of The Art of Electronics I'm really starting to get it. This book just has a way of making things understandable to anyone.

[PRR]

Horowitz and Hill, The Art of Electronics.

Note that there is very-new NEW Edition. Not cheap.

What he's added may be mostly stuff the Pedal World hasn't got into yet.

The older editions are VERY useful. Until recently they fetched good price on the used-book market. I guess the prices slid when the new edition came out?

IAC, the 1980s copies are VERY good buys, though a lot of reading.

[tubegeek]

New edition: 3rd. Gold/black cover.


Previous: 2nd. Silver/black cover.

Author's web site:
http://artofelectronics.net/

Closest thing we have to a One True Book.

[LightSoundGeometry]

https://www.google.com/search?q=the+art+of+electronics+pdf+download&oq=&aqs=chrome.1.69i58j0i66l3j5i66l2.2708341j1j4&sourceid=chrome&es_sm=0&ie=UTF-8

[plexi12000]

thank you guys- most of that is just plain over my head at this point.  but thanks for the reading suggestion!  --good lookin out light soundgeometry!

[hymenoptera]

Maybe trying to jump right into active stuff is what's confusing you. It really helps to understand passive components first. You probably already have a breadboard, some audio jacks, wire, and a selection of resistors and capacitors to choose from?

If you don't have a breadboard and a selection of resistors and capacitors, do that first, get those. Start tinkering with voltage dividers, and try to understand Ohm's Law. This will get you started on your way to understanding impedance.

Voltage dividers are where you use two resistors between two voltages, say for instance between signal and ground, and the current gets split at the point between those two resistors as a ratio of those two values, and thus the voltage that appears at that junction is based on that ratio. So for example the way we generate VRef (4.5v) from a 9v battery in pedals is by placing two of the same value resistors between 9v and ground, say two 100k resistors, and the voltage that appears between them is half the voltage difference (100k and 100k is a 50/50 ratio, and half of 9 is 4.5). You'll see this circuit in the PSU of many pedals, especially ones that use opamps. This same type of circuit can be used to attenuate signals. Two resistors between signal and ground, and at that junction, you'll get a voltage equal to the ratio of those two resistors.

Try to build some attenuator networks on the breadboard starting with simple 2 resistor networks, and put it between the guitar and amp. If built right it will quiet the guitar. A certain ratio of the signal at any given frequency gets "shunted" to ground, so it never appears at the next stage (the amp). Google image for "voltage divider" for examples.

Next, try replacing the second resistor (the one between signal and ground) with a potentiometer connected as a variable resistor (wiper lug strapped to one of the other lugs). This is exactly what's happening in your volume control in your guitar. The first resistor is like your pickups, the second is variable and shunts more/less signal to ground. As you change the resistance, the ratio of those two resistances changes.

Once you start understanding purely resistive networks, then try swapping a resistor with a capacitor. If you replace the first resistor (the series resistor) with a cap you get a high pass filter. If instead you replaced the second resistor (shunt resistor) with a cap, you get a low pass filter. This is what's happening with the tone control in your guitar! Did you know you can replace the capacitor in your guitar with another value and totally change the sound of that guitar?

Resistors don't care much about frequency, as least not until you get up into RF frequencies, and then weird shit starts happening. Capacitors block DC and low frequencies (DC is just zero hertz) and allow high frequencies to pass by coupling charges across their plates. Inductors are the opposite; they block high frequencies by absorbing the current into the magnetic field, but at low frequencies they act like a normal wire (which they are!). By combining resistors, capacitors, and/or inductors you can make complex voltage dividers where, for instance, a standard RC low pass filter (google "RC low pass") creates different impedance to the signal at different frequencies by shunting the higher frequencies to ground and letting the low ones pass unaffected. You could imagine low freqs going right by the cap liek it's not even there, but the poor highs fall through it like a trap door straight to ground! Now build this. It's just a couple parts.

Try some of these experiments on breadboard either between the guitar and amp, or between any other audio output and input, and try to build different reactive networks and try out some things. I know this isn't what you asked in the original post, but I think this is the best way to start to understand impedance, and I feel that understanding impedance is the key to grasping buffers.

Oh, and on the subject of Mr Ohm, just remember that 1 volt across 1 ohm, gives you 1 amp at 1 watt. Try to picture in your mind a 1 volt battery and a 1 ohm resistor. If you strap that 1 ohm resistor across those battery terminals, you'd get 1 amp of current. That's a lot for a little battery (don't try this!) and it will get hot. How hot? Well, the circuit will try to dissipate 1 watt of power! Remember that most resistors we use in pedals are only rated at 1-quarter watt! So it follows that the same 1 ohm resistor across a 9v battery (again DONT try this) would give you 9 amps! Of course a 9v can't actually give you 9 amps, but you can image it would try it's best and get pretty hot doing it. Everything we're building here on these forums and anywhere for that matter, all follow Ohm's Law, which is just the relationship between current (amperes) , potential (volts), and resistance (ohms). Ohms Law will help you understand impedance, and impedance will help you understand buffers. One thing at a time

[tubegeek]

Voltage dividers are where you use two resistors between two voltages, say for instance between signal and ground, and the current gets split at the point between those two resistors as a ratio of those two values, and thus the voltage that appears at that junction is based on that ratio.

Nitpicking here:

In a series circuit, the current is equal at all points in the circuit (Kirchoff's Voltage Law.) Current is not "split," but the voltage is.

Assumption for this application: no current is drawn from the IDEAL junction between the two resistors - in REAL circuits of course some current is usually drawn. I jumped in with this nitpick because the IDEAL voltage divider needs to be understood first.

The rest of the description is fine.

[MrStab]

when in a wider context of a complete guitar effect, and especially in the context of just starting out in DIY, i always think of input & output buffers like airlocks on a space station. don't take that at all literally. the input buffer "decontaminates" the astronaut (ie. lowers impedance) and prepares them for safe, efficient travel through the rest of the station. the output buffer prevents microbes from the astronauts spreading into space (other pedals), and prevents any decontaminated (competitively-low-impedance) aliens from coming in. xenophobes.

damn, that's a really bad analogy. far more credible than Mars One, though.
imagery like that lays the foundation for a more in-depth understanding for me, at least.

[hymenoptera]

Nitpicking here:

In a series circuit, the current is equal at all points in the circuit (Kirchoff's Voltage Law.) Current is not "split," but the voltage is.

Assumption for this application: no current is drawn from the IDEAL junction between the two resistors - in REAL circuits of course some current is usually drawn. I jumped in with this nitpick because the IDEAL voltage divider needs to be understood first.

The rest of the description is fine.

It's true, I could have worded much of that better, but it's hard sometimes to be both clear and technically correct I find. You're right that ideals must be understood before we delve into real world complexity. But I'm not sure it's clear to say that all currents are equal at all points in the circuit since this isn't a series circuit. That could be confusing as well. It might be proper to describe it as the current entering the junction is the same as the current leaving it, so if there's one entrance and two exits, then the current at the exits must be divided among them. That's how I understand Kirchoff's.

But you're right, it's volts that we're concerned with here, and I should have been more clear. Thanks for that

[plexi12000]

Thank you- thank-you!!!   good stuff.  i keep reading your answers...i think each time a tiny bit sinks in! lol

MrStab.....i get your analogy!! ha!   

I'm ok with components....i know what a voltage divider is.  some very basic stuff.  small bear elec. has a nice looking book about classic pedals -circuit theory.

forget the name offhand. sucker aint cheap.  but it looks pretty darn informative for a person at my level.

[tubegeek]

I'm not sure it's clear to say that all currents are equal at all points in the circuit since this isn't a series circuit.

Which is why I filled my nitpick with weasel words

If it WERE an ideal series circuit I would have been 100% correct though!

[hymenoptera]

Thank you- thank-you!!!   good stuff.  i keep reading your answers...i think each time a tiny bit sinks in! lol

awesome!

....i know what a voltage divider is.

okay, then imagine a voltage divider, two resistors right?

But imagine that while the shunt resistor is a fixed value, the series resistor is a different value depending on frequency. At low frequencies it's low resistance. The lower the frequency the more it starts look like a plain wire. At higher frequencies it gets higher resistance though. Go high enough it starts to look like an open circuit! Try to imagine how that would look at the output of the voltage divider if we're trying to push an audio signal through it. The ratio of series/shunt resistors will be different depending on the frequency. The attenuation will be frequency dependant! Imagine how that would sound?

If you imagined a low pass filter, you're right! This is basically what your pickup in your guitar is doing. Because the pickup is a coil of wire it is an inductor. The impedance is higher for high frequencies. So if you "load" the pickups with a relatively low shunt somehow, more of those high frequencies are shunted to ground compared to low frequencies. Use a high value shunt and there will be less of an effect.

When we talk about the input impedance (let's use the letter Z to describe impedance because it's just easier) of a pedal or amplifier preamp, etc, we mean how much does it load down the previous stage. Because guitars are so high Z (again, because of the thousands of windings needed to make a pickup) we have to be especially careful to have a high input Z at the next stage. And the inductive nature of the guitar makes this even more true, because a relatively low input Z at the next stage will have a greater attenuation of highs, and the guitar will sound muffled, just like when you roll down your tone knob.

A buffer is when the next stage has high enough input Z to not load down the previous stage too much, and low enough output Z to swing the voltage easily into whatever comes next. Well designed and implemented buffers try to pass the signal efficiently without having to worry too much about what comes before or after. Efficiency is the name of the game here.

That's all for now..

Some more terms to look up, "impedance bridging" and "impedance matching". And here's some good reading I found on this sort of stuff, an article at SoS:
http://www.soundonsound.com/sos/jan03/articles/impedanceworkshop.asp

I didn't read it all, but it looks like it gets pretty in depth and covers lots of audio related stuff that concerns us, as musicians, technicians, makers, and sound engineers.