Cornish Buffer - the value of values

[motterpaul]

Wow... just built a Cornish Buffer:


Looks easy and is except for one thing. I didn't have the specific R values in four out of nine cases. These days, it's rare for me to be missing any 1 component for almost anything I want to build here.

I know that with tolerances it's probably close enough to work, but still, when you add them all together, it (um...) adds up.  So, I went lower in each case.

50k = 47k
7.5k = 6.8k
20k = 18k
200k = 180k

I have another small vero, so I am going to build another and go the other way. I wonder if it will be an audible difference. And the diode - I have 1N5404s - and they are like barrels, the leads won't go thru the holes, so I used a good ol' 5817.

What do you guys think - is it just a matter of him using what was available to him? Or did he do it on purpose because those values really make a difference?

[Muadzin]

Why do you miss most of these values? With the exception of 50K I can get them all even from Tayda. And even then I can still get a 51K.

As for the weird values, I think with Pete Cornish you have to keep in mind the guy has a certain OCD. If something is tried and tested he will not deviate from it, no matter what! Take the G2. It has the Cornish bypass buffer, but as its a Muff, there is still some loading of the original signal. Does he deviate from his tried and tested buffer system and use something else? Like true bypass? Hell no! He put another buffer in between. Problem solved! And he's been doing it since the prehistory of rock. When certain odd values in caps and resistors were more readily available then they are now. So it makes sense that he keeps on using them.

Since you're building another version this could make for an excellent A/B test to see if it really matters. Battle of the Cornish buffers!

[motterpaul]

Well, I realize I can get them, but the interesting thing to me is to figure out why he used them. (just FYI, I just got started a year ago, and mainly do this to try to learn how these little varmints work, I am not pro designer or builder, nor do I have many years of acquiring components).

My point is that I have done maybe 30 builds here now, and I never needed the four odd values he selected. Between the 2 or 3 different "assorted collections" of resistors I started out with I have almost everything anyone ever listed. Some others I have special ordered (5k for ex.) because I have seen them more than once. But for resistors I have never encountered so many as a percentage of any one circuit where I did not have them in stock.

I am not complaining (blah, blah) - I am was just wondering if that is because the age of the design, or where he is located, or where he sources parts, or if he is just that specific about what he wants - and furthermore, how much difference will it make in the end for me to be that exact about it.

EDIT_ and to be clear, it looks like you answered the question, although other ideas are also welcome.

[Beaker]

Resistors come in a series of series' (is there such a thing?). Nowadays, and for most of the builds we use them for, E12 series represents the "standard" values. But there are also E24 and E48 series too.

Take a look at most of the BJFE builds, as Bjorn uses a LOT of E24 and sometimes E48 values. He probably bought thousands in a job lot once, and has been using them ever since. It's nice to have a assortment pack of E24 values, as they do come in handy sometimes.


http://circuitpages.com/wp-content/uploads/2014/10/EIA-Standard-resistor-values.gif

EDIT: you should be able to hit the "odd" values on this layout, using your existing E12 resistors, by wiring a couple in series. e.g. wire two 10k together to get the 20k.

[motterpaul]

Beaker - that is an excellent and very educational and informative answer for me. I have seen the terms E24, etc, but really just recently. Thank you so much, that makes a lot more sense now.

Perfect reply. Thanks again.

[motterpaul]

By the way - I do often get the exact values for caps & Rs by combining when I already have 99% of perfect components, but in this case I just wanted to try it quickly and I think it is a good experiment or me to try what I'm doing.

I would be interested in knowing how many other people have built this using E12 values, if they went higher or lower, and what they got out of it.

Also - I was reading a thing about "close" values and someone suggested that it is better to go one direction or the other in situations like this - because of tolerances, if you go low on one R, and high on another one close by, then you are going further away from the tolerance range and getting into more of a mismatch situation. That seems logical to me - if I explained it well enough.

I could draw it out with examples, but you all get the idea.

[Beaker]

Bjorn Juhls (BJFE) pedals use a lot of E24 values. I have seen on here people saying that his pedals need the correct values - they will work fine using nearest E12 values - just not as well.

I often wire two in series if I am missing values - looks ugly, but it works perfectly.

Search ebay for "E24 resistor assortment pack" or something similar. You can still get them quite easily.

[Beaker]

If I was going to build this, I would use these E12 values:

For 7k5   - 5k6 and 1k8 in series. ( Gives 7k4).
For 50k   - 22k and 27k in series. (Gives 49k).
For 20k   - two 10k  in series.
For 200k - two 100k in series.

[Frank_NH]

Paul - what do you think of the buffer?  Are you using it for something specific?  I built one thinking it would magically enhance my sound but I couldn't detect that it was doing anything.  Maybe I need to connect a 50 ft cord to my guitar...

[motterpaul]

Right now I am thinking about a replacement for the BC549 - I tried a BC109 - not bad. Tried a 2n5088 and pretty good.

But I am seeing that the only way to really evaluate this is to use it as a buffer to start a pedal chain. On its own all I hear is a slightly brighter, cleaner sound, but definitely very little gain, if any.

So far, I think the lower ohmage one sounds a little better. I might change those few resistors out and make them the right values (by combining like you said). The difference is very small, though, when comparing the two I have built on their own.

I am also going to buy the right values as a set like you said. There are other things I need to get anyway. Time to start one of those spreadsheets,  I think.

[motterpaul]

Yep, Frank. It doesn't do much at all on its own. Like you, I expected I would hear a pretty noticable difference but I don't. It does clean out the mud and enhance the treble in a nice, clear way. AND, if it makes your whole pedal chain happier (cleaner, clearer, brighter but not harsh) then that is what it is supposed to do.

Not a gain device though - seems more like something you would build permanently into an amp or maybe an EQ pedal. But my understanding is that it makes your whole "pedal chain" lower impedance and capacitance, which means the following pedals work better. I am going to put in my gig rig and see what happens, after a few mods like I was describing above.

I am thinking that in the right application it could make a nice difference. A lot of people seem to think so, anyway.

[Chris60601]

Paul,

I understand that some folks use it at the beginning of the chain and at the end. Your thoughts?

Cheers
Chris

[rocket88]

i've avoided this for awhile, as i've been choosing not to help you out paul for reasons i'm sure you know, but here goes. you need to understand what the role/job of a buffer is. if you're expecting there to be any sort of gain from using one, it's not going to happen. that's not it's job and it won't do it. a buffer is 1:1 as far as gain goes.

the job of a buffer is to take either a high capacitance input and make it a low capacitance output, or take a lower capacity input and raise the capacitance of the output. the idea is that when you plug a guitar into a cable you will loose some of the clarity and integrity of the signal based on the capacitance of the cable being higher then that of the pickups, there by loading your pickups, and as you have multiple pedals plugged into that cable you lose more and more signal quality. this is part or the problem when you use a ge fuzzface and a wah, if you place the fuzzface after the wah you'll get massive squeal and oscillation. also, this is part of  the reason for uber-expensive low capacitance cables, and without getting into an argument i can tell you there is a massive difference between cheap cables and some of the higher quality ones, both in tone, but also in volume. i wouldn't spend $100 dollars on a cable, but $30-50 i would after that some of the cables have too low capacitance and to me there's too much high end coming through. i also typically use 15-30ft cables no matter what, so longer cables = higher capacitance.

when you plug into a buffer, the buffer is supposed to remove the load on the pickups, and thereby regain the original sound of your instrument. now yes, some buffers are supposed to add a little something, a sparkle so to speak, like the klon buffer. but that may or may not be that noticeable on it's own. i have a soulfood, which has the ability to turn on and off the buffer, and you will notice the difference, but the klon buffer on it's own, i don't know how much you will notice. another example is with the cornish buffer and the G2. without the buffer at the front end of the G2, it will sound like utter shit, put it in front and the thing sounds fantastic and magical.

remember, that it's not so much about one thing being magic and making your instrument sound great, but its the sum of the parts, and what your gear is. not to pick on TGP people, but you don't want to be one of them that things ___________ pedal will make you sound like hedrix, that's the magic.

[Ciaran Haslett]

@ Chris

I've built 2 into the same box on my pedaltrain.  The idea was to buffer the start AND the end of my chain cos thats what I read and people where raving about this buffer.

In practise....nearly all my fuzz pedals hate the damn thing!  My meathead and FF types become a thin raspy mess.  I actually spent some time troubleshooting my Fulltone 70 thinking I'd screwed up somewhere until I powered it up on its own on the bench.  Night and Day!

My Muffesque builds are a lot more forgiving bit there's still a hint of extra presence that I don't really like.

Now YMMV depending on your pups, amp, cable length.  My main rig is a Strat with SCNs into a Deluxe Rvb (sold the Twin thank god) and a pedaltrain with 10 trueBP pedals.  Cables are 25ft total.  I can barely hear any capacitance loss going straight into the amp vs through the pedalboard so, as far as using a buffer is concerned, I don't want one.  Out it came!

I still have the one at the end of the chain simply because the box they are in makes a handy patch bay for having both my guitar/amp cables on the same side.

[Ciaran Haslett]

Incidentally Rocket

Thats probably why both G2s i put together were muddy dark hallions!  This was before I built the buffer.  Might have to hunt them out again.

[Beaker]

"if you're expecting there to be any sort of gain from using one, it's not going to happen. that's not it's job and it won't do it. a buffer is 1:1 as far as gain goes. "

Yep, that's the truth all right - a buffer will not give any gain / volume boost at all - It's just not in their job description.

Ciaran is correct too - buffers will annihilate any old school fuzz, distortion, some wahs and some modulation pedals, if a  buffer, or buffered pedal is put before it.

I made a Death By Audio Supersonic fuzz Gun for a customer, explicitly telling him not to put it after a buffered pedal, and the reasons why. "Yeah OK, no problem" he said. Two hours later, he was on the phone telling me it wasn't working properly. When I asked him where in his pedal chain he had put it, it turned out he had about four buffered pedals in front of it. I reminded him what I said about not putting it after a buffer, and he replied with "But that's where it needs to go".

Some people just don't take a hint!

On the other hand a buffer at the end of your signal chain can sometimes work very well. Many pro pedal boards come with switchable buffers at front and end, so you can elect to have them on or off, depending on pedal usage.

[Beaker]

BTW I have built several buffers, and the paranoid sceptic in me always worried whether they were doing anything at all, or was I imagining hearing it.

I was about to sling out an old damaged and very long  power extension cable, that got trashed when my old drummer ran over it in his van. Instead, I soldered a jack socket to each end, and use it between a newly built buffer and my amp.

That really shows if the buffer works right or not!

[rocket88]

Exactly. A with good buffer you won't even know if it's on.

Cian, the reason a lot of old school fuzzes had buffers in in their design. Fuzzfaces for instance work by loading the effect. Once you change that it goes to hell. So what you experienced would be expected. With the Cornish buffer and the g2, I would pretty much guarantee that if you put the buffer in front of it it'll come to life. I think it was Javi, who kept testing it and figured it out so I can't take credit for that one.

[Ciaran Haslett]

Just went over the G2 thread here and found Javi's comments.  Yip...it absolutely needs the buffer.  I also found some comments of my own involving attempts to make it sound right.  Apparently I was initially happy with my results

Must have been Builders Euphoria!  Definitely gonna throw the buffer in front of one when I find it.  IF I find it!

Good man Rocket

[rocket88]

nice one man. there are a bunch of my builds that i fell in love with when i first built them, then i recently plugged them in and feel i have to make it better. hell, i recently pulled apart my modified bass rat. i used to love the 1n270 diodes, but recently found them to no have enough grit, so i spent a few hours yesterday swapping diodes till i found what i wanted, not to mention i now have a super high gain metal can lm308h in there instead of a normal one, and there's a massive difference. makes me wonder if the change in IC took away the magic of the 1n270 diodes i had in there before.

anyways, back to the topic at hand about impedance. i once had impedance explained to me like by thinking about a garden hose with water hooked up to a bottle. apparently Mr. Black had heard the same explanation because on their blog it has the same explanation, which is:

Play this scenario out with me:
We have a signal coming out of our guitar.  Since we've got guitar pickups in our guitars, not nuclear steam turbines, generating the signal (current), we've got a small signal.  That is to say there isn't a whole ton of current backing our signal.  Its weak.  But it is strong enough to make our amplifiers work, and here's where it gets fun.

Lets say our guitar is connected to a circuit with a low input impedance.  The signal can flow in freely right?  Right!  But everything at its cost.  The signal flows so freely into the input that the input actually starts to put demand on the signal, making it even smaller!

Imagine your guitar cable is a hose carrying your signal.  There is a changing pressure in the hose.  This changing pressure is your signal.  The hose is connected to a big bottle, with contstant low pressure inside.  Lower pressure than what is in the hose.  Nature being natural, everything wants to seek equilibrium and enlightenment, so the pressure in the hose moves rapidly to fill the big bottle.  This lowers the pressure in the hose and begins to put a demand on whatever is supplying the pressure, as if to say:  GIMME MORE PRESSURE!
This is the "loading effect" we want to avoid, so to speak.  At least with a buffer, or current amplifier.

Now play this scenario out:
We have the same signal coming out of our guitar.  The same weak signal.
We connect our guitar to a circuit with a high input impedance.  Now the signal can't flow so freely into the circuit.  It would follow that we are just getting a weak little signal into our amplifier, right?  Well, you are right, but we don't need the whole thing.  In fact, we don't need any of the actual current.  All we need is the voltage; the change in pressure.  Not the actual pressure.
Because there is no decrease (loading) in the changing pressure in the hose, we can see the full change of pressure, or the full signal.

Hopefully, these examples help to illustrate why we want a high input impedance (or resistance to current) in our buffer circuit.

hopefully this will help.