FX "always on"

[soulsonic]

Okay, I looked at the Pete Cornish website, and I see what he's on about with the buffering and whatnot. I can understand the whole thing about the buffer circuits and running everything low-Z. That is actually a very practical approach when dealing with complex live setups. If you're interested in a Cornish-style switching system, based on using your LD-2, drop me an email and I may be able to suit your needs for far less than $18,000.

[Mark Hammer]

Troll?

No.  Just expectations mismatched to the responses received, in tandem with being forced to work in a 2nd language.

[Ben N]

Ah, well, just the same, I could do without the "entitlement" attitude and demand forinstant gratification.

[Mark Hammer]

As someone who regularly has to struggle to get out "big ideas" in a second language, I have to say it is far too easy to sound pompous, impatient, and a whole bunch of other less than flattering qualities when working in a second language that one is imperfectly fluent in.  I'll happily give the gentleman the benefit of the doubt and hope he's lurking around long enough to realize he's more welcome than he thought.

[themartin]

Ben N

troll maybe but not deaf at all...

s*&?%

[ADR]

If you are planning to use a cornish buffer at the front of a chain, and possibly at the end before the cable run to your amp, then a true bypass strip makes a lot of sense.

The buffer will present a consistent input impedance to your guitar, and drive whatever effects you insert after it cleanly.

What many here are suggesting, is that the relationship between input impedance and guitar should be roughly 10-to-1, as Cornish recommends. (i.e. if your output impedance from your guitar is 100KOhm, then you want 1MOhm input impedance on the next device it plugs into). This is pretty standard practice in a lot of designs you'll find here.

When you true-bypass an effect that is located after such a buffer, then the guitar will still see that 10:1 relationship, and the buffer will drive whatever afterwards in its own low output Z, hopefully low enough to drive whatever input impedances on the next device cleanly. If you have only output switching for the various effects, the load from the circuit inputs are still in the chain and they change their input impedances to the buffer output to being less than this 10:1 ratio.

Quoted from Cornish's No TrueBypass article on his website:

"My system, which I devised in the early 70's, is to feed the guitar into a fixed high impedance load, which is identical to the amp input, and then distribute the signal to the various effects and amps by low impedance buffered feeds. "

This is more or less having a buffer at the front that is always in the signal chain, followed by effects, and then buffered out, which is what I outlined above. What I suspect Cornish does, is if an effect is bypassed, the effect circuit (and it's i/o buffers) are removed from the signal path! When an effect is engaged, the buffers before and after the effect (which he installs) maintain that consitent impedance relationship, regardless of the effect circuit design with possible deviations from ideal input and output impedances. It's a sledgehammer approach for a flyswatter. i.e. It's guaranteed to work.

Now what happens if you don't remove these from the signal path when not in use is that you have buffers going into more buffers in your signal path, and this is the Boss approach, which can yield cumulative noise and possible signal degredation. The best way is to have only as few buffers as required in the signal path at any one point.

The easy way to do this is with, tada, a true bypass strip. Assuming the circuits are well designed, it should be sufficient to have one buffer right after the guitar located at the front of the pedalboard, then your pedals via the bypass strip, and then another buffer last in the chain to drive the cable going to the amp. You then insert other buffers as required for pesky effects that have impedance ratio mismatches in relationship to whatever it follows. This has the least amount of modifications, will yeild clean signals, and properly drive your cables, while still adhering to the Cornish principles. What Cornish does is more or less the same thing, but on every effect that is called in, has his own (Cornish) buffers sandwiching the effect, and this is what I refer to as the sledgehammer approach. Chances are with most modern effects, you won't have to install any sandwiching buffers, (as some like Boss already have their own). It's only on old tonesucking output-only switching designs that this would be required.

If you read more of the articles listed on geofex.com and muzique.com, it will become more clear as to what is really needed to make this stuff work cleanly.

I don't need any FX loop or true by-pass. I don't want any of that in my gtr rig.

I simply want to solder all my cabling from pedal-to-pedal, solder the DC power and bypass the output of each FX from a DPDT foot switch so that way from my gtr to the amp it will be the same impedance with or without FX in the chain. All is buffered by Pete Cornish LD-1 or LD-2.

Thanks

[soulsonic]

No, I get it. The idea of using a low impedance buffer at the beginning of the signal chain means you can have a chain of pedals with only the outputs bypassed, and the low impedance created with all the inputs of the pedals won't suck the tone (like would normally happen) because the buffer (LD-2) is perfectly happy with the total impedance. What this means is that since the inputs of the effects are always in the circuit, then the impedance of the entire chain is constantly the same, and therefore the guitar and amp will respond consistently regardless of what effects are switched in and out. Using a true bypass system defeats this, so that's why it's important NOT to use true bypass if you're trying to do a constant impedance system like the one Pete Cornish advocates. I see the truth of it, and I understand what it's about - it seems like exactly the sort of solution that is usually necessary for large professional setups - serious setups with very long cables. I don't think he actually has to install buffers in the effects themselves - this is taken care of by the buffer that drives the effects chain.

The only problem is that you absolutely HAVE to use a buffer designed to drive very low impedances or the setup won't work. Martin said he already has the required buffer box, so the rest of the setup should be easy.

[ADR]

The key is having that buffer(s) at the beginning and end of the effects chain to break up the cable length. IMO, that's the biggest accomplishment of a properly buffered signal. I noticed the difference right away after installing my tillman cable preamp. It was quieter, clearer, and generally better responding volume controls on my guitar. The rest of my pedals also behaved much better. At the end of the day, that's the desired result I wanted. If anything, I think a signal should be buffered right after the pickups. But on my hollowbodies, this isn't practical, hence the tillman with the preamp inside the guitar's plug.

After the initial buffer, I don't believe that it makes all that much of a difference whatever switching scheme the subsequent devices in the chain use as long as the initial buffer's output Z is low enough to drive everything cleanly, and you have something on the tail end to make the distance to the amp.

I'm still scratching my head why people have such hesitation towards using TB  switching w/ some buffers, but I'm not gonna defend this tooth and nail. It works for me, I'd thought I'd help out and share my experiences and help out. Honestly, I haven't had any evil tone destroying moments using some TB switching in my rig, so I thought it's a valid option.

Martin should be able to make a bypass strip, but instead of truebypass, use output bypass only. Why bother modifying the pedals if you don't have to?

[soulsonic]

ADR, I think we're in agreement here. The only case where I can see the possibility that the strict "Cornish-style" setup could be distinct advantage would maybe in a system with lots of pedals and VERY long cables.

[ADR]

I've tested my tillman 20' cable with an additional three daiseychained 20' XLR's for a total of 80', with no tonal differences. 90% of the time, the single 20 footer is more than enough lead for my live work. Oh ya, there's another 20' cablesnake from the pedalboard buffer out to the spaceecho by the amp, then the amp. I only have 4 effects on my board and a tuner, so it isn't a big whopper, but again, no problems kicking stuff in and out of the signal path. Knock on wood, etcetera etcetera. If I can drive 80' of cable, the small amount of cable it has to navigate in a bypass strip is negligable.

I think Cornish spec'd his cable mini driver to work up to 100' of cable from guitar to board. I don't know of many applications where one would need all that, but hey, it's there, and the tillman is more or less the same idea and has similar performance.

If there was one device that made my whole time reading this board so worthwhile, it's the Tillman. Thank you Don, where ever you are.

[Ben N]

What this means is that since the inputs of the effects are always in the circuit, then the impedance of the entire chain is constantly the same, and therefore the guitar and amp will respond consistently regardless of what effects are switched in and out.

Well, no, it still isn't always the same, because bypassed pedals inputs are in parallel; active pedals are in series. So, for example, in a four pedal setup, assuming each pedal has a 500k input-Z and 50k output-Z, the master input buffer has an output-Z of 1k, and no master output buffer: When all the pedals are bypassed, their inputs are effectively all in parallel with the amp input (assume 1M), so the buffer is seeing a combined load of 111.11k, which easily meets the 10 to 1 rule. But now, what if only pedal #1 is kicked in: the buffer sees a 500k load, which is well and good; but now the pedal's 50k output is seeing a combined load of  142.86K--definitely a recipe for tone-sucking. With the first two pedals in line, the combined load is still 200k, against that same 50k.

Of course, my assumptions here about numbers may be all off--I honestly don't know and welcome correction. But my point is that to get practically consistent impedance through the chain, you have to:
EITHER buffer each pedal fore and aft (like Boss/DOD)--
OR use TBP and a master input buffer and an output buffer on each pedal. As for buffering the outputs, I suspect that this is what Pete/Martin means by "well-designed", i.e. pedals with consistent behavior in the signal chain because they are buffered. Be that as it may, doing it this way guarantees that your signal is buffered where it has to be, but not repetitively or unnecessarily (as in the Boss approach of "buffer 'em all, and let God sort 'em out"), since the individual output buffers get bypassed with their effects.

Ben

[shredgd]

No, I get it. The idea of using a low impedance buffer at the beginning of the signal chain means you can have a chain of pedals with only the outputs bypassed, and the low impedance created with all the inputs of the pedals won't suck the tone (like would normally happen) because the buffer (LD-2) is perfectly happy with the total impedance. What this means is that since the inputs of the effects are always in the circuit, then the impedance of the entire chain is constantly the same, and therefore the guitar and amp will respond consistently regardless of what effects are switched in and out. Using a true bypass system defeats this, so that's why it's important NOT to use true bypass if you're trying to do a constant impedance system like the one Pete Cornish advocates. I see the truth of it, and I understand what it's about - it seems like exactly the sort of solution that is usually necessary for large professional setups - serious setups with very long cables. I don't think he actually has to install buffers in the effects themselves - this is taken care of by the buffer that drives the effects chain.

The only problem is that you absolutely HAVE to use a buffer designed to drive very low impedances or the setup won't work. Martin said he already has the required buffer box, so the rest of the setup should be easy.

Assuming this is right and good, you would still have a major problem: you won't be able to use more than one pedal at once, or you will have your effects in parallel (I know, it might work for a chorus, but...)

[shredgd]

After building my pedals strictly true bypass, I've happily converted myself to pre and post-pedalboard buffering, too.

You get the best of both worlds! Your cleans will still be sparkling (dull cleans is mostly the problem with long cables without buffering), but you won't have the additional noise because of a dozen or more buffers in your signal chain.
And I suggest a simple bipolar buffer as first buffer (i.e. in front of your pedalboard) because it is the one which more faithfully reproduces the original sound and response of your guitar.

[themartin]

Thanks Soulsonic for putting in english words what is in my head,

finally glad people are getting it but please let's get back at what this forum started for.............FX always On.

Enough about TBP or not.........this isn't why we're here. My mind is set.

I simply want to know if someone could help me figure out a way to keep my FX always ON when i power my rig..........

- Digitech whammy WH-1 (momentary switching)
- Boss CE-2 (momentary switching)
- MXR original script phase 90........i'm guessing this one is always ON.........i'm still searching for the OK schematics (can't find 'em)
- Boss LS-2 line selector (momentary switching)............which i use for a buffered FX loop in my rig for an Echoplex EP-4 with 40' of cables (might become 60' later on)

Total cable run...........120' and might become 150, or 160' later.........thinking of adding a Cornish NB-2 Booster + Line Driver.

Got to go now........thanks again for your time.

Martin...........who wasn't all that wrong after all.

[slacker]

What this means is that since the inputs of the effects are always in the circuit, then the impedance of the entire chain is constantly the same, and therefore the guitar and amp will respond consistently regardless of what effects are switched in and out.

Well, no, it still isn't always the same, because bypassed pedals inputs are in parallel; active pedals are in series.

Damn you Ben, that's exactly what I'd been planning to say, it was probably going to be my longest ever post, Mark Hammer would have been proud

To get back to the original question to make Boss pedals always on have a look at this thread http://www.diystompboxes.com/smfforum/index.php?topic=58678.0 again everything I was planning to say has already been covered.
I'm pretty sure old MXR pedals use latching stomps to do non truebypass so to make it always on just stomp the switch so the pedal is on and leave it there. 

[Ben N]

Damn you Ben, that's exactly what I'd been planning to say, it was probably going to be my longest ever post, Mark Hammer would have been proud

Sorry, insomnia. 

+1 on the always-on thing. Anyway, it would appear that the only pedal on Martin's board that might present a loading issue is the Phase 90 (I have an original block-letter, so I am familiar) since everythng else is presumably well-buffered.

[soulsonic]

There's a discussion going on here: http://www.diystompboxes.com/smfforum/index.php?topic=58678.0   where they are talking about this same issue. It looks like RG Keen is going to be writing an article about how this can be done. The Boss pedals should be simple enough, but I'm not sure what circuit the Whammy uses for bypass, either way, it can be worked out.

If the Phase 90 causes problems, it would be very easy to add a simple buffer to the end of it so it won't foul the constant impedance system.

Are you planning on removing the the pedals from their boxes and mounting them in the pedalboard the way Cornish does?

[themartin]

hopping so.......

But i might have to have him build my rig, can't find all the infos i need.

For now i've tested and set all this :

- PSU - OK
- Floorboard construction done (3/4" russian plywood, aluminium plates, sheilding) - OK
- FX units - have been the same since 8-9 years (will have an optionnal FX loop in the pedalboard for studio use)

I really only have to know how to keep my FX on as soon as they receive 9 or 18 vdc when i power the thing.

thanks again

Martin